System

Abstract

A system is provided.SOLUTION: A system comprising: means for tapping an item in which a viewer is interested while viewing moving image content on a device; means for transmitting tapped position information and frame data to a server; means for analyzing an object in a frame based on the transmitted information and specifying a corresponding item; means for searching a database based on feature information of the specified item and acquiring related product information; and means for transmitting the acquired product information to the device and displaying the product information to the viewer.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In recent years, video content has rapidly become popular, and viewers are increasingly interested in the performers and items in the videos. However, there are limited ways for viewers to instantly obtain information about items that interest them, which often results in missed purchase opportunities. Furthermore, existing methods make it difficult to identify items in videos, and even if they are identified, the information provided to viewers is insufficient. There is a need for a system that can resolve these issues and allow viewers to easily access items of interest. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides the following means. The system includes: a means for allowing a viewer to tap on an item of interest while watching video content on a device; a means for transmitting the tapped location information and frame data to a server; a means for analyzing an object in the frame based on the transmitted information and identifying the relevant item; a means for searching a database based on characteristic information of the identified item to obtain related product information; and a means for transmitting the obtained product information to the device and displaying it to the viewer. The system also includes a means for ranking related product information based on the viewer's past browsing history and purchase history; a means for providing the viewer with optimal product information based on the ranked information; a means for accepting interactions by the viewer to view detailed information from the provided product information; and a means for transitioning to an e-commerce site to display the detailed information. This allows viewers to easily obtain detailed information about items of interest, which can lead to subsequent purchasing behavior.

[0006] A "viewer" is a user who watches video content.

[0007] A "device" is an electronic device that plays video content and accepts user interaction.

[0008] "Video content" refers to video data such as movies, dramas, and advertisements that are played by viewers.

[0009] An "item" is a part of an object or person that appears in video content and is of interest to viewers.

[0010] A "tap" is an action in which the viewer lightly touches the device screen with their finger.

[0011] "Location information" refers to the specific coordinate data of the location where the viewer tapped on the device screen.

[0012] "Frame data" is still image data that represents a specific moment in video content.

[0013] The "server" is a remote computer system that receives and analyzes tapped position information and frame data.

[0014] "Analysis" is the process of processing data and extracting specific information.

[0015] An "object" is a specific item that can be recognized within video content.

[0016] "Feature information" is detailed data such as color, shape, and pattern that can be used to identify an object.

[0017] A "database" is a system that stores and manages product information and other related data.

[0018] "Searching" is the process of locating specific information within a database.

[0019] "Product information" refers to detailed product data related to an item in which the user is interested, such as price and a link to a sales site.

[0020] "Past browsing history" refers to data about products that a viewer has previously purchased or viewed.

[0021] "Purchase history" is detailed data on products that a viewer has purchased in the past.

[0022] "Ranking" is the process of ranking multiple items based on specific criteria.

[0023] "Interaction" refers to the operations or actions that viewers perform on a device.

[0024] An "EC site" is a website that conducts electronic commerce. [Brief explanation of the drawings]

[0025] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0026] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0027] First, the terms used in the following description will be explained.

[0028] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0029] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0030] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0031] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0032] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0033] [First embodiment]

[0034] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0035] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0036] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0037] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0038] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0039] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0040] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0041] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0042] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0043] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0044] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0045] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0046] The present invention provides a system that allows viewers to easily obtain information about items they are interested in while watching video content on their device, leading to purchasing behavior. This system starts when the viewer taps on an item they are interested in on the device screen, and includes a process in which a server recognizes the item and provides related product information.

[0047] Server-side processing

[0048] 1. Running the video analysis module:

[0049] The server receives the video content and begins analyzing it frame by frame, using an object recognition AI model to identify items within the frame.

[0050] 2. Database search of identification results:

[0051] Based on the information of the recognized object (for example, color, shape, brand logo, etc. for clothing), the e-commerce site's database is searched to identify related product information.

[0052] 3. Ranking of specific products:

[0053] Based on the search results, the AI ​​ranks the most suitable products based on the viewer's past browsing and purchasing history.

[0054] 4. Provision of Information:

[0055] Information about the tapped item is instantly sent to the device, and data is prepared that links to the details page of the product the viewer needs.

[0056] Terminal side processing

[0057] 1. Listening for tap events:

[0058] The device monitors the viewer's tap events and sends the location information to the server when a specific item is tapped.

[0059] 2. Sending a request to the server:

[0060] A request is sent to the server along with the frame data of the tap position, and this information is used by the server's analysis module.

[0061] 3. Information Receipt and Display:

[0062] It receives product information sent from the server and displays related information on the screen while the user is viewing the content. It improves user convenience by displaying links to related e-commerce sites and thumbnail images of the products.

[0063] User processing

[0064] 1. Item tap:

[0065] The viewer taps on an item that interests them (e.g., the clothes worn by an actress). This action records the location where the tap was made on the device.

[0066] 2. Information verification:

[0067] Check the product information displayed on your device. Click on the link to the related e-commerce site to view more information.

[0068] 3. Purchasing decision:

[0069] They decide on their purchasing behavior based on the information provided, and if they like the product, they complete the purchase process through the e-commerce site.

[0070] Specific examples

[0071] 1. User Scenario:

[0072] While watching a drama starring an actress on his smartphone, a user becomes interested in the dress the actress is wearing and taps on the dress.

[0073] 2. Tap event occurs:

[0074] The device detects the tap event and sends the tap position and frame information to the server.

[0075] 3. Parsing on the server:

[0076] The server analyzes the items in the frame and recognizes the dress. Based on the recognition results, it searches the e-commerce site database to identify the most suitable dress.

[0077] 4. Information provision:

[0078] The server sends the details of the dress and a purchase link to the device, which then displays the details of the dress to the user based on the received information.

[0079] 5. Purchasing behavior:

[0080] The user checks the displayed information, and if they like the dress, they click the provided link to go to the e-commerce site and complete the purchase process.

[0081] In the above-described form, the present invention allows viewers to quickly and easily obtain information about items that interest them, thereby promoting purchasing behavior.

[0082] The processing flow will be explained below.

[0083] Step 1:

[0084] A user watches video content on a device and taps on an item of interest (e.g., an actress's clothing).

[0085] Step 2:

[0086] The device detects the touch event and retrieves the touch location and the current video frame data, including a specific timestamp in the video.

[0087] Step 3:

[0088] The device sends the tap location, frame data, and timestamp to the server, along with any necessary metadata.

[0089] Step 4:

[0090] Based on the tap position information and frame data received by the server, a video analysis module is used to analyze the object in the corresponding frame.

[0091] Step 5:

[0092] The server uses the analysis to identify the item closest to the tapped location, which is performed by an object recognition AI model.

[0093] Step 6:

[0094] The server extracts the characteristic information of the identified item (e.g., color, shape, brand, pattern) and searches the e-commerce site's database.

[0095] Step 7:

[0096] The server retrieves relevant product information from the database and ranks the most suitable products by referring to the viewer's past browsing and purchasing history.

[0097] Step 8:

[0098] The server sends the ranked product information (product name, price, sales site link, image, etc.) to the terminal.

[0099] Step 9:

[0100] The product information received by the device is displayed on the screen in a pop-up or overlay format.

[0101] Step 10:

[0102] The user checks the displayed product information and clicks on the link to the product detail page of the product that interests them.

[0103] Step 11:

[0104] The device will redirect to the EC site, where the user can view detailed information and then complete the purchase procedure.

[0105] Example 1

[0106] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0107] When viewers watch video content on their devices, there is a demand for a system that can quickly and easily obtain information about items that interest them and directly link this information to purchasing behavior. Conventional systems have the problem that the process from finding a product that interests a viewer to purchasing it is complicated and time-consuming. Furthermore, they lack a mechanism for suggesting optimal products based on the viewer's past preferences and browsing history.

[0108] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0109] In this invention, the server includes means for analyzing objects in the frame based on the transmitted information and identifying the relevant item, means for searching a database based on attribute information of the identified item and acquiring related product information, and means for transmitting the acquired product information to the device and displaying it to the viewer. This enables a viewer to quickly acquire product information related to an item that interests them in the video, and receive appropriate product suggestions, simply by tapping on that item.

[0110] A "viewer" is someone who watches video content on a device and interacts with items of interest.

[0111] A "device" is a device that a viewer uses to watch video content, and includes, for example, a smartphone, a tablet, a PC, and the like.

[0112] "Video content" means media, including video and audio, that is played by a viewer on a device.

[0113] "Item" refers to an object or subject that appears within video content.

[0114] "Tapping means" refers to a device or method by which a viewer operates the touchscreen of a device to select a particular item.

[0115] "Tap location information" refers to the coordinate information when a viewer taps an item on the device screen.

[0116] "Frame data" refers to image data of a specific moment (frame) in video content.

[0117] "Server" refers to a computer system that receives and processes data sent from a viewer's device.

[0118] "Means for transmitting" refers to a method or apparatus for communicating data between a device and a server.

[0119] "Means for analyzing" refers to a device or method for identifying an object in video content based on the transmitted information and extracting its attribute information.

[0120] "Means for identifying" refers to a device or method that identifies the relevant item from the analyzed data.

[0121] "Attribute information" refers to characteristic information about the identified item (e.g., color, shape, brand logo, etc.).

[0122] The term "means for searching a database" refers to a device or method for obtaining related product information from a database based on the attribute information of a specified item.

[0123] "Product Information" refers to detailed information about related products (e.g., price, image URL, purchase link, etc.).

[0124] "Means for transmitting and displaying to the viewer" refers to a method or apparatus for transmitting product information obtained from the server to the viewer's device and displaying it on the viewer's screen.

[0125] "Past browsing history" and "purchase history" refer to historical information about content that a viewer has viewed or products that they have purchased in the past.

[0126] The term "ranking means" refers to a device or method that evaluates and prioritizes relevant product information based on the viewer's past browsing history and purchasing history.

[0127] The "means for accepting interaction" refers to a method or device that allows a viewer to perform an operation to view detailed information from the provided product information.

[0128] "Means for transitioning to an e-commerce site" refers to a device or method that provides a link to an e-commerce site and directs the viewer to that site for more information.

[0129] The present invention provides a system that allows viewers to easily obtain information about items they are interested in while watching video content on their device, leading to purchasing behavior. This system starts when the viewer taps on an item they are interested in on the screen of their device, and includes a process in which a server recognizes the item and provides related product information.

[0130] Server-side processing

[0131] The server receives video content sent from the viewer's device and analyzes the video frame by frame using an object recognition AI model (e.g., YOLOv5 or Mask R-CNN). The server identifies the items present in each frame and extracts their attributes (e.g., color, shape, brand logo).

[0132] Based on the attributes of the identified objects, an SQL query is executed to search the e-commerce site's database to retrieve matching related products. The server also uses machine learning models (e.g., collaborative filtering and deep learning) to rank the best products based on the viewer's past browsing and purchasing history.

[0133] The server constructs the identified product information and ranking results in JSON format and sends them to the terminal using a RESTful API.

[0134] Terminal side processing

[0135] The device uses a touchscreen event listener to monitor events when a specific item in the video is tapped. The touch location information is captured and temporarily saved.

[0136] This location information and the current frame data are sent to the server using an HTTP POST request. The product information received from the server is analyzed and relevant information is displayed on the screen while the user is watching. Product information includes thumbnail images, prices, purchase links, etc.

[0137] User processing

[0138] Viewers tap on an item of interest (for example, clothing worn by an actress in a drama). The location of the tap is recorded on the device and sent to the server. The viewer checks the product information sent from the server and clicks on a link to the related e-commerce site to view more information. If they find a product they like, they can complete the purchase process through the e-commerce site.

[0139] Specific examples

[0140] For example, a user might be watching a TV drama on their smartphone and become interested in the dress an actress is wearing. When the user taps on the dress, the device detects the tap event and sends the tap location and frame information to the server. The server then analyzes the items in the frame and recognizes the target dress. Based on the recognition results, the database of the e-commerce site is searched to identify the most suitable dress.

[0141] The server sends detailed information about the dress and a link to purchase it to the device, which then displays it to the viewer. The user checks the information displayed, and if they like the dress, they click the provided link to go to the e-commerce site and complete the purchase process.

[0142] This system allows viewers to quickly and easily obtain information about items that interest them, thereby promoting purchasing behavior.

[0143] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0144] Step 1:

[0145] Monitor tap events (device)

[0146] The device monitors events when the viewer taps on a specific item in the video, and uses a touchscreen event listener to obtain the location information when the viewer taps on the item.

[0147] Input: Viewer taps.

[0148] Data processing: Obtain the coordinates of the tapped location and temporarily save them.

[0149] Output: Coordinate information of the tap position.

[0150] Specific behavior:

[0151] Set a touch event listener on the video player.

[0152] Get the tapped coordinate position and save it in a variable.

[0153] Step 2:

[0154] Send tap information to server (device)

[0155] The device sends the acquired tap position information and the current frame data to the server using an HTTP POST request.

[0156] Input: Tap location and frame data.

[0157] Data processing: Convert tap position information and frame data into JSON format.

[0158] Output: The HTTP POST request sent to the server.

[0159] Specific behavior:

[0160] Build tap position information and frame data in JSON format.

[0161] Creates an HTTP POST request and sends it to the server.

[0162] Step 3:

[0163] Video analysis (server)

[0164] The server analyzes the objects in the frame based on the received tap information and frame data, and uses an object recognition AI model (e.g., YOLOv5 or Mask R-CNN) to identify the items and extract their attribute information.

[0165] Input: Tap position information and frame data.

[0166] Data processing: Input the frame data into an object recognition AI model to extract specific item attributes (color, shape, brand logo, etc.).

[0167] Output: Attribute information of the recognized item.

[0168] Specific behavior:

[0169] The frame data stored on the server is input into the object recognition AI model.

[0170] Temporarily save the attribute information of the item obtained from the model.

[0171] Step 4:

[0172] Database search (server)

[0173] The server searches the e-commerce site's database based on the attribute information of the identified item to retrieve related product information. It executes an SQL query to retrieve matching product information from the database.

[0174] Input: Item attribute information.

[0175] Data processing: Generate SQL queries and retrieve product information from the database.

[0176] Output: Retrieved product information.

[0177] Specific behavior:

[0178] Generates an SQL query using the item's attribute information.

[0179] Query the database to retrieve related product information.

[0180] The acquired product information is temporarily saved.

[0181] Step 5:

[0182] Product information ranking (server)

[0183] The server ranks the product information obtained based on the viewer's past browsing and purchasing history, and uses machine learning models (e.g., collaborative filtering and deep learning) to select the best products.

[0184] Input: Retrieved product information, viewer's past browsing and purchasing history.

[0185] Data processing: Input data into a machine learning model and rank it.

[0186] Output: Ranked product information.

[0187] Specific behavior:

[0188] Obtain past viewer data from the database.

[0189] Product information and historical data are input into a machine learning model to obtain ranking results.

[0190] Temporarily save the ranking results.

[0191] Step 6:

[0192] Product information sent to server (server)

[0193] The server sends the ranked product information to the terminal and returns the product information in JSON format using a RESTful API.

[0194] Input: Ranked product information.

[0195] Data processing: Build product information in JSON format.

[0196] Output: The HTTP response sent to the device.

[0197] Specific behavior:

[0198] Build product information in JSON format.

[0199] Use the RESTful API to generate an HTTP response and send it to the device.

[0200] Step 7:

[0201] Display product information (device)

[0202] The terminal analyzes the product information received from the server and displays related information on the screen being viewed.

[0203] Input: The HTTP response from the server.

[0204] Data processing: Analyze JSON data and extract product information.

[0205] Output: Product information displayed on the screen.

[0206] Specific behavior:

[0207] Parse the JSON data received from the server.

[0208] Display the extracted product information on the screen (e.g., by placing a thumbnail image or purchase link).

[0209] Step 8:

[0210] Check detailed information (user)

[0211] The user checks the product information displayed on the terminal and clicks on the link to view more detailed information.

[0212] Input: View product information.

[0213] Data processing: Clicking on a link.

[0214] Output: Show detailed information.

[0215] Specific behavior:

[0216] Check the detailed information of the displayed product.

[0217] Click on the link to access the e-commerce site, if desired.

[0218] Step 9:

[0219] Purchasing decision (user)

[0220] The user decides on a purchase based on the information provided, and if they find a product they like, they complete the purchase process through the e-commerce site.

[0221] Input: Display product details.

[0222] Data processing: Operations to carry out purchase procedures.

[0223] Output: Product purchase completed.

[0224] Specific behavior:

[0225] Check the product details and decide to purchase.

[0226] Proceed through a purchase on an e-commerce site.

[0227] (Application example 1)

[0228] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0229] In today's content distribution services, it is difficult for viewers to quickly and accurately obtain information about items they are interested in while watching a video, leading to immediate purchase behavior. There is also a need for a system that allows viewers to obtain detailed information about items along with related products. Given this situation, a means is needed to efficiently satisfy viewers' interests and encourage purchasing behavior without compromising the viewing experience.

[0230] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0231] In this invention, the server includes means for tapping on an item of interest to a viewer while watching video content on a device, means for transmitting tapped position information and frame data to the server, means for analyzing each video frame using an object recognition AI model to identify the relevant item, means for searching a database based on feature information of the identified item to obtain related product information, means for transmitting the obtained product information to the device and displaying it to the viewer, and means for generating detailed data about the tapped item as a prompt sentence using a generative AI model and presenting it to the viewer. This makes it possible to quickly obtain information about the item being viewed and promote purchasing behavior by providing necessary detailed information while maintaining the viewing experience.

[0232] A "viewer" is a user who watches video content.

[0233] A "device" is a hardware device for viewing video content, and specifically includes smartphones, tablets, and computers.

[0234] "Video content" refers to content that combines video and audio that is viewed by a viewer through a device.

[0235] An "item" is an object or element that appears in video content and is an object that may interest viewers.

[0236] A "tap" is an action in which the viewer touches the device's touch panel and identifies the location.

[0237] "Location information" refers to the coordinate data on the screen of the device that was tapped.

[0238] "Frame data" is still image data that represents a specific moment in video content.

[0239] A "server" is a computer system that receives, processes, and analyzes data sent from devices.

[0240] An "object recognition AI model" is an artificial intelligence algorithm for identifying specific items within a frame of video content.

[0241] "Feature information" is data relating to the attributes and identifiers of the identified item.

[0242] A "database" is a collection of information in which related product information is stored.

[0243] "Obtained product information" is detailed data about products related to the identified item.

[0244] A "generative AI model" is an artificial intelligence algorithm that generates detailed data about a specific item.

[0245] A "prompt" is a detailed explanatory text generated by a generative AI model to be presented to the viewer.

[0246] "Viewing history" is a record of video content that a viewer has viewed in the past.

[0247] "Purchase history" is a record of products that a viewer has purchased in the past.

[0248] "Ranking" is an evaluation process that ranks related products based on viewing history and purchasing history.

[0249] "Interaction" refers to the operations or inputs that viewers make through a device.

[0250] An "EC site" is a website for conducting electronic commerce.

[0251] The present invention provides a system for promptly and accurately providing information about items of interest to viewers while they are watching video content on their devices, thereby promoting purchasing behavior. An embodiment of this system will be described below.

[0252] System configuration

[0253] This system mainly consists of a server, viewer terminals, and a database.

[0254] 1. Server

[0255] Hardware: General cloud server (e.g. AWS EC2)

[0256] Software: Flask (web framework), OpenCV (image processing library), TensorFlow or PyTorch (object recognition AI model), generative AI model

[0257] Function: The server receives tap events from viewers, analyzes video frames to identify items, retrieves relevant product information from a database, and generates detailed prompts using a generative AI model.

[0258] 2. Terminal

[0259] Hardware: Smartphones, tablets

[0260] Software: Tap event monitoring application, HTTP request library

[0261] Function: The device detects tap events while the viewer is watching video content, sends the location information and frame data to the server, and displays the retrieved product information.

[0262] 3. Database

[0263] Components: A database containing product information for an e-commerce site

[0264] Function: Provides related product information based on the characteristics of the identified item.

[0265] Operation flow

[0266] 1. Viewer Controls

[0267] When a viewer is watching video content on a device, they tap on an item they are interested in. At this time, the device transmits information about the tap position and the corresponding frame data to the server.

[0268] 2. Processing on the server

[0269] Based on the received data, the server uses an object recognition AI model to analyze the items in the frame and identify the relevant item. It then searches a database based on the identified item's characteristics (e.g., color, shape, brand) to retrieve related product information. It then uses a generative AI model to generate a detailed prompt for the relevant product and present it to the viewer. This prompt includes the product name, price, and a purchase link.

[0270] 3. Display on device

[0271] The product information received from the server is displayed on the device, and the viewer can check the provided information. If the viewer is interested, they can be redirected to an e-commerce site for more detailed information or to complete the purchase process.

[0272] Specific examples

[0273] While watching a drama on a smartphone, a viewer becomes interested in the blue dress worn by the actress and taps on that part. At the same time, the device sends its location information and frame data to a server. The server uses an object recognition AI model to analyze the items in the frame and identify the "blue dress." It then searches a database to retrieve related product information (product name: "Blue Dress," price: "10,000 yen," purchase link, thumbnail image). It then uses a generative AI model to generate the following prompt:

[0274] "While watching the video, I noticed a blue dress the actress was wearing. When I tapped on the item, more information was displayed on the screen."

[0275] Viewers can check the product information on their smartphones based on the information sent from the server, and then click on the link to be taken to the e-commerce site and complete the purchase process.

[0276] The above is the "Mode for Carrying Out the Invention" of the present invention. This system allows viewers to quickly and accurately obtain information about items that interest them, and promotes purchasing behavior.

[0277] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0278] Step 1:

[0279] Viewers tap on items that interest them while watching video content on their device.

[0280] Input: Viewer taps

[0281] Output: Tap location and frame data of the video being watched

[0282] How it works: The device captures and records the location of the tap and the video frame data at that moment. This information is used to make requests to the server.

[0283] Step 2:

[0284] The tapped position information and frame data are sent to the server.

[0285] Input: Tap location and frame data of the video being watched

[0286] Output: HTTP request to the server

[0287] Operation: The device generates and sends an HTTP request to the server to send the collected location information and frame data.

[0288] Step 3:

[0289] The server analyzes the objects in the frame based on the information received and identifies the relevant item.

[0290] Input: Tap location and frame data

[0291] Output: Feature information of the identified item (e.g. color, shape, brand)

[0292] How it works: The server uses an object recognition AI model to analyze the frame data and identify the item of interest based on the transmitted tap location. From this analysis, it extracts feature information about the item.

[0293] Step 4:

[0294] A database is searched based on the characteristic information of the identified item to obtain related product information.

[0295] Input: Feature information of the identified item

[0296] Output: Related product information (product name, price, purchase link, thumbnail image, etc.)

[0297] Operation: The server uses the acquired feature information to search the database to identify and retrieve relevant product information. The search results include the most relevant product information.

[0298] Step 5:

[0299] Based on the generated product information, a prompt sentence is generated using a generative AI model and presented to the viewer.

[0300] Input: Related product information

[0301] Output: Prompt (e.g., "While watching the video, I noticed the blue dress the actress was wearing. If you tap on the item, more information will appear on the screen.")

[0302] How it works: The server uses a generative AI model to convert the acquired product information into detailed prompts that are used to communicate product details to the viewer.

[0303] Step 6:

[0304] The server sends product information and a prompt to the device.

[0305] Input: Prompt statement and product information

[0306] Output: Sending data to the viewer's device

[0307] How it works: The server generates a response to send the generated prompt and product information to the viewer's device, and sends it to the device using the HTTP protocol.

[0308] Step 7:

[0309] The terminal displays the received product information and prompt text on the screen.

[0310] Input: Received product information and prompt text

[0311] Output: Product information and prompt text displayed on the screen

[0312] Operation: The terminal analyzes the product information and prompt text received from the server and displays it on the screen in a format that is easy for the viewer to see.

[0313] Step 8:

[0314] Viewers click on the provided link to view more information.

[0315] Input: Product information link

[0316] Output: Transition to EC site and display of detailed information

[0317] How it works: When a viewer clicks on a product link displayed on the screen, the device's browser will redirect to the corresponding e-commerce site, where they can view detailed information and complete the purchase process.

[0318] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0319] The present invention is a system that easily obtains information about items that interest a viewer while they are watching video content on their device, and by combining this with an emotion engine, provides information and promotes purchasing behavior that is tailored to the viewer's emotions. This system starts when the viewer taps on an item of interest on the device screen, and includes a process in which a server recognizes the item and provides related product information. The system also uses the emotion engine to analyze the viewer's emotions and optimize the information provided.

[0320] Server-side processing

[0321] 1. Running the video analysis module:

[0322] The server receives the video content and begins analyzing it frame by frame, using an object recognition AI model to identify items within the frame.

[0323] 2. Database search of identification results:

[0324] Based on the information of the recognized object (for example, color, shape, brand logo, etc. for clothing), the e-commerce site's database is searched to identify related product information.

[0325] 3. Ranking of specific products:

[0326] Based on the search results, the AI ​​ranks the most suitable products based on the viewer's past browsing and purchasing history.

[0327] 4. Running the Emotion Engine:

[0328] The server analyzes the viewer's facial expressions and voice data using an emotion engine to identify the viewer's emotional state.

[0329] 5. Emotion-based filtering:

[0330] Filter or rank product information based on the viewer's emotional state.

[0331] 6. Provision of Information:

[0332] Information about the tapped item is instantly sent to the device, and data is prepared that links to the details page of the product the viewer needs.

[0333] Terminal side processing

[0334] 1. Listening for tap events:

[0335] The device monitors the viewer's tap events and sends the location information to the server when a specific item is tapped.

[0336] 2. Sending a request to the server:

[0337] A request is sent to the server along with the frame data of the tap position, and this information is used by the server's analysis module.

[0338] 3. Information Receipt and Display:

[0339] It receives product information sent from the server and displays related information on the screen while the user is viewing the content. It improves user convenience by displaying links to related e-commerce sites and thumbnail images of the products.

[0340] User processing

[0341] 1. Item tap:

[0342] The viewer taps on an item that interests them (e.g., an actress's clothing). This action records the location where the tap was made on the device.

[0343] 2. Information verification:

[0344] Check the product information displayed on your device. Click on the link to the related e-commerce site to view more information.

[0345] 3. Purchasing decision:

[0346] They decide on their purchasing behavior based on the information provided, and if they like the product, they complete the purchase process through the e-commerce site.

[0347] Specific examples

[0348] 1. User Scenario:

[0349] While watching a drama starring an actress on his smartphone, a user becomes interested in the dress the actress is wearing and taps on the dress.

[0350] 2. Tap event occurs:

[0351] The device detects the tap event and sends the tap position and frame information to the server.

[0352] 3. Parsing on the server:

[0353] The server analyzes the items in the frame and recognizes the dress. Based on the recognition results, it searches the e-commerce site database to identify the most suitable dress.

[0354] 4. Running the Emotion Engine:

[0355] The server analyzes the user's emotions from facial expressions and voice data and determines, for example, whether the user is excited.

[0356] 5. Emotional information:

[0357] Based on the information from the emotion engine, the server provides preferentially information on products that suit the user's state of excitement.

[0358] 6. Information provision:

[0359] The server sends the details of the dress and a purchase link to the device, which then displays the details of the dress to the user based on the received information.

[0360] 7. Purchasing behavior:

[0361] The user checks the displayed information, and if they like the dress, they click the provided link to go to the e-commerce site and complete the purchase process.

[0362] According to the above-described embodiment, the present invention can quickly and easily obtain information about items that interest a viewer, and can further promote purchasing behavior by displaying information according to the viewer's emotions.

[0363] The processing flow will be explained below.

[0364] Step 1:

[0365] A user watches video content on a device and taps an item that interests them (e.g., an actress's clothing). This action records the location of the tap on the device.

[0366] Step 2:

[0367] The device detects a touch event and obtains the touch location, the current video frame data, and a timestamp, including the time at which the operation occurred.

[0368] Step 3:

[0369] The device sends the tap location, frame data, and timestamp to the server in a request that also includes the necessary metadata.

[0370] Step 4:

[0371] Based on the tap location information and frame data received by the server, a video analysis module is used to analyze the objects in the frame, using an object recognition AI model to identify the item closest to the tapped location.

[0372] Step 5:

[0373] The server extracts the characteristic information (e.g., color, shape, brand, pattern) of the identified item and searches the e-commerce site's database to obtain related product information.

[0374] Step 6:

[0375] The server retrieves relevant product information from the search results and ranks the most suitable products based on the viewer's past browsing and purchasing history, taking user preferences into account in the process.

[0376] Step 7:

[0377] The server runs an emotion engine that analyzes the viewer's facial and voice data to determine their emotions, in this case emotional states such as excitement, joy, and surprise.

[0378] Step 8:

[0379] The server further filters the ranked product information based on the user's emotional state, prioritizing products that best suit the user's current emotions.

[0380] Step 9:

[0381] The server sends the identified product information (product name, price, sales site link, image, etc.) to the device. This information is optimized based on emotions.

[0382] Step 10:

[0383] The product information received by the device is displayed on the screen in a format that is intuitive and easy for viewers to understand, including pop-ups and overlays.

[0384] Step 11:

[0385] Users review the displayed product information and click on the link to the product detail page of the product they are interested in. This action makes it easier to obtain additional information.

[0386] Step 12:

[0387] The device will redirect to an e-commerce site where the user can view detailed information, allowing them to make a final purchasing decision.

[0388] Example 2

[0389] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0390] Conventional video viewing systems lack a means for viewers to quickly and easily obtain information about items that interest them within video content. Furthermore, there is no method for providing information or encouraging purchasing behavior based on the viewer's emotional state. This makes it difficult to effectively stimulate viewers' purchasing desire and provide an optimal purchasing experience.

[0391] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for tapping an item in which a viewer is interested, a means for transmitting tapped position information and frame data to the server, a means for analyzing an object in the frame based on the transmitted information and identifying the relevant item, a means for searching a database based on characteristic information of the identified item and acquiring related product information, a means for analyzing facial expressions and voice data of the viewer and identifying the emotional state, and a means for filtering or ranking product information based on the identified emotional state. This makes it possible to quickly and easily acquire information about items in which the viewer is interested, and to optimally display information and promote purchases according to the viewer's emotions.

[0392] A "viewer" is a user who is viewing video content on a device.

[0393] A "device" is an electronic device (e.g., smartphone, tablet, computer) that displays video content and can be operated by a user.

[0394] "Video content" means digital media containing a combination of moving images and sounds that convey information through visual and auditory means.

[0395] An "item" refers to an object or person that appears in video content and is an object of interest to viewers.

[0396] "Tap" means that the viewer selects an item by touching the device's touchscreen.

[0397] "Tap location information" is the coordinate data when a viewer taps an item on the device screen.

[0398] "Frame data" refers to information about each frame (still image) of video content, and is data that represents video at a specific point in time.

[0399] A "server" is a central processing unit that receives data from devices and performs analytical processing and database searches.

[0400] "Object analysis means" refers to the technology or algorithms used by the server to identify objects or items present in each frame of video content (e.g., object recognition AI models).

[0401] The "means for identifying the relevant item" is a method for finding the item that the viewer tapped on from among the analyzed objects.

[0402] The "means for searching a database" is a method for querying a database in which various product information is stored based on the characteristic information of the identified item.

[0403] "Related Product Information" is detailed information (e.g., price, description, images, links) about products that are similar or match the identified item.

[0404] The "means for displaying to the viewer" refers to a method for visually presenting the acquired product information on a device.

[0405] "Means for analyzing facial and / or audio data" means the techniques or algorithms used to infer the emotional state of a viewer from their facial expressions or audio (e.g., emotion engine).

[0406] "Emotional state" is data that reflects the type and intensity of a viewer's emotion, including joy, excitement, sadness, etc.

[0407] A "filtering or ranking means" is a method for sorting or prioritizing retrieved product information based on the identified emotional state.

[0408] The present invention is a system that allows a viewer to quickly and easily obtain information about items that interest them while watching video content on a device, and provides information that corresponds to the viewer's emotions and promotes purchasing behavior.

[0409] The overall operation of the system involves three entities: the server, the device, and the viewer. We will explain how each entity interacts and processes information.

[0410] Server-side processing

[0411] First, we will explain the server-side processing. Specifically, it has the following functions:

[0412] 1. Video Analysis Module:

[0413] The server splits the video content received from the user into frames and identifies items within the frames using an object recognition AI model (e.g., YOLO or Detectron2). Through this analysis, the server identifies items that may be of interest to the viewer.

[0414] 2. Database Search:

[0415] Based on the analyzed item's attribute information (color, shape, brand logo, etc.), the server queries the e-commerce site's database (e.g., Amazon Product Advertising API, Rakuten product database) to obtain related product information.

[0416] 3. Ranking Algorithm:

[0417] Based on the acquired product information, the server takes into account the viewer's past browsing and purchasing history and uses AI algorithms (such as collaborative filtering or content-based filtering) to rank the products that are best suited to the viewer.

[0418] 4. Sentiment Analysis Module:

[0419] The server acquires the viewer's facial expression and voice data and analyzes it using an emotion engine (e.g., Microsoft Azure Emotion API, IBM Watson Tone Analyzer) to determine their real-time emotional state (excitement, joy, calmness, etc.).

[0420] 5. Emotion-based filtering:

[0421] The server filters or re-ranks product information based on the viewer's emotional state, for example, prioritizing special offers and limited edition products for an excited viewer.

[0422] 6. Provision of Information:

[0423] The server immediately transmits the filtered product information to the terminal and provides a link to the detailed page of the product the viewer needs.

[0424] Terminal side processing

[0425] Next, the processing on the terminal side will be described.

[0426] 1. Listening for tap events:

[0427] The device monitors the viewer's tapping on the screen while watching a video, acquires the location information of the tap, and sends this location information to the server.

[0428] 2. Data transmission:

[0429] The tapped position information and frame data are sent to the server, which uses this information in the server's video analysis module.

[0430] 3. Information Receipt and Display:

[0431] It receives product information sent from the server and displays related information on the screen while the user is watching. Specifically, it displays product information pop-ups, links to related e-commerce sites, thumbnail images of the products, and more.

[0432] User processing

[0433] Finally, the processing on the user side will be described.

[0434] 1. Item tap:

[0435] The user taps on an item of interest (e.g., the actress's clothing) in the video they are watching. This action records the location of the tap on the device.

[0436] 2. Information verification:

[0437] The user checks the product information displayed on the device, clicks on the link to the related e-commerce site, and views the product details page.

[0438] 3. Purchasing decision:

[0439] The user decides on a purchasing decision based on the information provided, and if they like the product, they complete the purchase process through the e-commerce site.

[0440] Specific examples

[0441] Next, a specific example of the operation of the present invention will be described.

[0442] 1. User Scenario:

[0443] While watching a drama starring an actress on his smartphone, a user becomes interested in the dress the actress is wearing and taps on the dress.

[0444] 2. Tap event occurs:

[0445] The device detects the tap event and sends the tap position and frame information to the server. Specifically, it sends the tap coordinates and the frame data at that time.

[0446] 3. Parsing on the server:

[0447] The server analyzes the items in the received frame and recognizes the dress. Based on the recognition results, it searches the e-commerce site database to identify the most suitable dress.

[0448] 4. Running the Emotion Engine:

[0449] The server analyzes the user's emotions from facial expressions and voice data, and determines, for example, whether the user is in an excited state.

[0450] 5. Emotional information:

[0451] The server provides information on products that suit the user's state of excitement with priority based on information from the emotion engine.

[0452] 6. Information provision:

[0453] The server sends detailed information about the dress and a link to purchase it to the device, which then displays the details of the dress to the user based on the information received.

[0454] 7. Purchasing behavior:

[0455] The user checks the displayed information, and if they like the dress, they click on the provided link to go to the e-commerce site and complete the purchase process.

[0456] Example prompt:

[0457] "If I want to know about the dress an actress is wearing, how can the server provide this information and explain the process for the user to purchase it?"

[0458] As described above, the system of the present invention is designed to quickly and easily obtain information about items that interest viewers, and to display information optimally in response to the viewers' emotions and promote purchases.

[0459] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0460] Step 1:

[0461] Listening for touch events

[0462] The device monitors the viewer's tapping on the screen while watching a video. The input is the viewer's tapping on the touchscreen. Specifically, the device obtains the tapped position information (coordinate data) and the frame data indicating which frame was being displayed. The output is the tapped position information and frame data.

[0463] Step 2:

[0464] Sending a request to the server

[0465] The device sends the acquired tap position information and frame data to the server. The inputs include tap position information and frame data. Specifically, the device sends this information to the server in the form of an HTTP request or similar. The output includes the tap position information and frame data sent to the server.

[0466] Step 3:

[0467] Running the video analysis module

[0468] The server uses the received frame data to run an object recognition AI model (e.g., YOLO or Detectron2) and analyze the objects in the frame. The input is the frame data. Specifically, the server analyzes the frame and identifies the item. The output is information about the analyzed item.

[0469] Step 4:

[0470] Database search of identification results

[0471] The server searches the e-commerce site's database based on the analyzed item's characteristic information (color, shape, brand logo, etc.). The analyzed item information is the input. Specifically, the server calls the API to retrieve related product information. The retrieved product information is the output.

[0472] Step 5:

[0473] Ranking of specific products

[0474] The server uses AI algorithms (such as collaborative filtering or content-based filtering) to rank the acquired product information. The inputs are the acquired product information and the viewer's past browsing history and purchase history. Specifically, the server uses AI to rank the products. The output is the ranked product information.

[0475] Step 6:

[0476] Running the Emotion Engine

[0477] The server acquires the viewer's facial expression and voice data and analyzes it using an emotion engine (e.g., Microsoft Azure Emotion API, IBM Watson Tone Analyzer). The input is the viewer's facial expression and voice data. Specifically, the server analyzes the emotion data and identifies the viewer's emotional state. The output is the analyzed emotional state.

[0478] Step 7:

[0479] Sentiment-based filtering

[0480] The server filters or re-ranks the product information based on the analyzed emotional state. The inputs include the analyzed emotional state and the ranked product information. As a specific operation, the server sorts or prioritizes the product information according to the emotional state. The output includes the filtered or re-ranked product information.

[0481] Step 8:

[0482] Providing information

[0483] The server transmits the filtered or re-ranked product information to the terminal. The input includes the filtered or re-ranked product information. As a specific operation, the server converts the product information into a data packet and transmits it to the terminal. The output includes the product information transmitted to the terminal.

[0484] Step 9:

[0485] Information reception and display

[0486] The device receives product information sent from the server and displays related information on the screen being viewed. The input is the product information sent from the server. In concrete terms, the device presents the product information to the user in the form of a pop-up display or related links. The output is the product information displayed in a format that the viewer can check.

[0487] Step 10:

[0488] Information verification and purchasing behavior

[0489] The user checks the product information displayed on the device and, if interested, proceeds with the purchase. The input is the product information displayed on the device. The specific operation is for the user to click on a link to the relevant e-commerce site and proceed with the purchase. The output is a result indicating whether the purchase was completed.

[0490] Through the above steps, viewers can easily obtain information about items that interest them, and information is displayed according to their emotions, promoting purchasing behavior.

[0491] (Application example 2)

[0492] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0493] Currently, there are insufficient ways for viewers to easily obtain information about items they are interested in while watching video content.In addition, information is not provided according to the viewer's emotional state, making it difficult to suggest optimal product information.

[0494] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0495] In this invention, the server includes means for tapping on an object of interest to a viewer while watching video content on a device, means for transmitting tapped position information and frame data to the server, means for analyzing objects in the frame based on the transmitted information and identifying the relevant object, means for searching a database based on characteristic information of the identified object and acquiring related product information, means for transmitting the acquired product information to the device and displaying it to the viewer, means for analyzing the viewer's facial expressions and voice data using an emotion engine and identifying the viewer's emotional state, and means for filtering or ranking the product information to be provided based on the viewer's emotional state. This allows the viewer to quickly and easily obtain information on items that interest them while watching a video, and further makes it possible to provide optimal product information according to the viewer's emotional state.

[0496] A "viewer" is a person who watches video content.

[0497] A "device" is an electronic device that displays video content and allows viewers to interact with it.

[0498] "Video content" refers to all media content provided in video format.

[0499] An "object" is an object that a viewer becomes interested in while watching video content and tries to obtain information by tapping on it.

[0500] A "tap" is the act of touching a specific location on a device screen once with a finger or stylus.

[0501] "Frame data" refers to each still image data that constitutes video content.

[0502] A "server" is a computer system that receives a request over a network, executes the specified process, and returns a response.

[0503] "Analyzing an object" means recognizing a specific object from frame data in video content and extracting information about it.

[0504] "Feature information" refers to attribute information for identifying an object, such as the object's color, shape, brand logo, etc.

[0505] A "database" is a collection of digital information organized to allow efficient storage and retrieval of specific information.

[0506] "Product Information" refers to detailed information about products and services related to the subject matter.

[0507] An "emotion engine" is a software module that analyzes the viewer's emotional state and optimizes the information provided based on the results.

[0508] "Emotional state" refers to data that represents the viewer's current psychological and physiological state.

[0509] "Filtering" refers to the act of sorting and prioritizing the information provided.

[0510] "Ranking" is the act of assigning a rank to information or objects based on certain criteria.

[0511] The present invention is a system that easily obtains information about objects of interest to viewers while they are watching video content on their devices, and provides optimal product information according to the viewer's emotional state. The system for implementing the present invention consists of a server, a terminal, and user interactions.

[0512] Server Processing

[0513] The server consists of the following modules:

[0514] Object Recognition Module: Using TensorFlow and OpenCV, this module analyzes and identifies objects within frames of video content. This module recognizes objects based on the transmitted location information and frame data, and extracts their feature information.

[0515] Database search module: Using MySQL or MongoDB, this module searches the database for relevant product information based on the characteristics of the identified object. This module quickly retrieves and ranks product information.

[0516] Sentiment analysis engine: Using Google Cloud Vision and Amazon Rekognition, the engine analyzes the viewer's emotional state from facial expressions and voice data, and filters or ranks the product information provided based on the analysis results.

[0517] Terminal handling

[0518] The terminal has the following features:

[0519] Monitoring tap events: Events that occur when a viewer taps the screen of the device are monitored, and the location information and frame data are sent to the server.

[0520] Communication with the server: When a tap event is detected, the information is sent to the server as a request and product information sent from the server is received.

[0521] Information display: Product information obtained from the server is displayed on the screen while the content is being viewed, and related product links and thumbnail images are provided.

[0522] User Action

[0523] The user interacts with:

[0524] Item Tap: Tap on an object that interests you while watching and send its location information.

[0525] Check information: Check the product information displayed on your device and click on the link for more information or to purchase.

[0526] Purchasing behavior: Select a product based on the information provided and complete the purchase process on the e-commerce site.

[0527] Specific examples

[0528] For example, suppose a user is watching a movie on a media distribution platform and is interested in the clothing worn by the lead actor. In this case, the user taps on the device screen. This tap event is detected by the device, and the relevant location information and frame data are sent to the server. The server identifies the clothing based on this data and searches for related product information from a database based on its feature information. Furthermore, an emotion analysis engine analyzes the user's emotional state (e.g., excitement or curiosity), ranks the product information accordingly, and sends it to the device. The device displays the results to the user, who can click on the provided link to check more information and make a purchase.

[0529] Prompt Sentence Examples

[0530] While a user is watching a movie on their smartphone, they tap on an item they are interested in. Can you give us an example of a system that retrieves information about that item and provides relevant product information, taking into account the user's emotional state?

[0531] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0532] Step 1:

[0533] User interaction and tap events

[0534] While a user is watching video content on a device, they tap on an object that interests them.

[0535] Input: User tap position information and video frame data

[0536] What happens: The device's touchscreen detects the user's taps and captures their location.

[0537] Step 2:

[0538] Sending tap position

[0539] The terminal transmits the tap position information and the associated frame data to the server.

[0540] Input: User tap position information and video frame data

[0541] Output: Tap position information and frame data are sent to the server.

[0542] Specific operation: The device detects a touch event and sends the coordinates and current frame data to the server as an HTTP request.

[0543] Step 3:

[0544] Analysis of objects in the frame

[0545] The server analyzes the objects in the frame based on the transmitted location information and frame data, and identifies the relevant target.

[0546] Input: Tap position information and video frame data

[0547] Output: Feature information of identified objects

[0548] Specific operation: The object recognition module (TensorFlow or OpenCV) on the server recognizes the object around the tap position in the frame and extracts its feature information.

[0549] Step 4:

[0550] Database search

[0551] The server searches the database based on the characteristic information of the identified object and obtains related product information.

[0552] Input: Feature information of the identified object

[0553] Output: List of product details

[0554] Specific operation: Using the feature information, query the database (MySQL or MongoDB) to retrieve the corresponding product information.

[0555] Step 5:

[0556] Emotional state analysis

[0557] The server uses an emotion analysis engine to identify the viewer's emotional state based on the user's facial expression data and voice data.

[0558] Input: User's facial expression data and voice data

[0559] Output: Emotional state assessment result

[0560] Specific operation: Analyzes the user's emotional state using APIs such as Google Cloud Vision and Amazon Rekognition.

[0561] Step 6:

[0562] Filtering and ranking product information

[0563] The server filters or ranks the product information based on the analyzed emotional state.

[0564] Input: List of product information, emotional state evaluation results

[0565] Output: Filtered or ranked product information

[0566] Specific behavior: Select and prioritize the most relevant product information based on emotional state.

[0567] Step 7:

[0568] Sending and Displaying Information

[0569] The server transmits the filtered product information to the terminal, which displays it on the screen being viewed.

[0570] Input: Filtered or ranked product information

[0571] Output: Display information on the user's device (product image, link to details, etc.)

[0572] Specific operation: The server sends the generated product information back to the device as an HTTP response, and the device receives the information and displays it on the screen.

[0573] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0574] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0575] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0576] [Second embodiment]

[0577] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0578] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0579] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0580] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0581] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0582] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0583] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0584] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0585] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0586] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0587] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0588] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0589] The present invention provides a system that allows viewers to easily obtain information about items they are interested in while watching video content on their device, leading to purchasing behavior. This system starts when the viewer taps on an item they are interested in on the device screen, and includes a process in which a server recognizes the item and provides related product information.

[0590] Server-side processing

[0591] 1. Running the video analysis module:

[0592] The server receives the video content and begins analyzing it frame by frame, using an object recognition AI model to identify items within the frame.

[0593] 2. Database search of identification results:

[0594] Based on the information of the recognized object (for example, color, shape, brand logo, etc. for clothing), the e-commerce site's database is searched to identify related product information.

[0595] 3. Ranking of specific products:

[0596] Based on the search results, the AI ​​ranks the most suitable products based on the viewer's past browsing and purchasing history.

[0597] 4. Provision of Information:

[0598] Information about the tapped item is instantly sent to the device, and data is prepared that links to the details page of the product the viewer needs.

[0599] Terminal side processing

[0600] 1. Listening for tap events:

[0601] The device monitors the viewer's tap events and sends the location information to the server when a specific item is tapped.

[0602] 2. Sending a request to the server:

[0603] A request is sent to the server along with the frame data of the tap position, and this information is used by the server's analysis module.

[0604] 3. Information Receipt and Display:

[0605] It receives product information sent from the server and displays related information on the screen while the user is viewing the content. It improves user convenience by displaying links to related e-commerce sites and thumbnail images of the products.

[0606] User processing

[0607] 1. Item tap:

[0608] The viewer taps on an item that interests them (e.g., the clothes worn by an actress). This action records the location where the tap was made on the device.

[0609] 2. Information verification:

[0610] Check the product information displayed on your device. Click on the link to the related e-commerce site to view more information.

[0611] 3. Purchasing decision:

[0612] They decide on their purchasing behavior based on the information provided, and if they like the product, they complete the purchase process through the e-commerce site.

[0613] Specific examples

[0614] 1. User Scenario:

[0615] While watching a drama starring an actress on his smartphone, a user becomes interested in the dress the actress is wearing and taps on the dress.

[0616] 2. Tap event occurs:

[0617] The device detects the tap event and sends the tap position and frame information to the server.

[0618] 3. Parsing on the server:

[0619] The server analyzes the items in the frame and recognizes the dress. Based on the recognition results, it searches the e-commerce site database to identify the most suitable dress.

[0620] 4. Information provision:

[0621] The server sends the details of the dress and a purchase link to the device, which then displays the details of the dress to the user based on the received information.

[0622] 5. Purchasing behavior:

[0623] The user checks the displayed information, and if they like the dress, they click the provided link to go to the e-commerce site and complete the purchase process.

[0624] In the above-described form, the present invention allows viewers to quickly and easily obtain information about items that interest them, thereby promoting purchasing behavior.

[0625] The processing flow will be explained below.

[0626] Step 1:

[0627] A user watches video content on a device and taps on an item of interest (e.g., an actress's clothing).

[0628] Step 2:

[0629] The device detects the touch event and retrieves the touch location and the current video frame data, including a specific timestamp in the video.

[0630] Step 3:

[0631] The device sends the tap location, frame data, and timestamp to the server, along with any necessary metadata.

[0632] Step 4:

[0633] Based on the tap position information and frame data received by the server, a video analysis module is used to analyze the object in the corresponding frame.

[0634] Step 5:

[0635] The server uses the analysis to identify the item closest to the tapped location, which is performed by an object recognition AI model.

[0636] Step 6:

[0637] The server extracts the characteristic information of the identified item (e.g., color, shape, brand, pattern) and searches the e-commerce site's database.

[0638] Step 7:

[0639] The server retrieves relevant product information from the database and ranks the most suitable products by referring to the viewer's past browsing and purchasing history.

[0640] Step 8:

[0641] The server sends the ranked product information (product name, price, sales site link, image, etc.) to the terminal.

[0642] Step 9:

[0643] The product information received by the device is displayed on the screen in a pop-up or overlay format.

[0644] Step 10:

[0645] The user checks the displayed product information and clicks on the link to the product detail page of the product that interests them.

[0646] Step 11:

[0647] The device will redirect to the EC site, where the user can view detailed information and then complete the purchase procedure.

[0648] Example 1

[0649] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0650] When viewers watch video content on their devices, there is a demand for a system that can quickly and easily obtain information about items that interest them and directly link this information to purchasing behavior. Conventional systems have the problem that the process from finding a product that interests a viewer to purchasing it is complicated and time-consuming. Furthermore, they lack a mechanism for suggesting optimal products based on the viewer's past preferences and browsing history.

[0651] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0652] In this invention, the server includes means for analyzing objects in the frame based on the transmitted information and identifying the relevant item, means for searching a database based on attribute information of the identified item and acquiring related product information, and means for transmitting the acquired product information to the device and displaying it to the viewer. This enables a viewer to quickly acquire product information related to an item that interests them in the video, and receive appropriate product suggestions, simply by tapping on that item.

[0653] A "viewer" is someone who watches video content on a device and interacts with items of interest.

[0654] A "device" is a device that a viewer uses to watch video content, and includes, for example, a smartphone, a tablet, a PC, and the like.

[0655] "Video content" means media, including video and audio, that is played by a viewer on a device.

[0656] "Item" refers to an object or subject that appears within video content.

[0657] "Tapping means" refers to a device or method by which a viewer operates the touchscreen of a device to select a particular item.

[0658] "Tap location information" refers to the coordinate information when a viewer taps an item on the device screen.

[0659] "Frame data" refers to image data of a specific moment (frame) in video content.

[0660] "Server" refers to a computer system that receives and processes data sent from a viewer's device.

[0661] "Means for transmitting" refers to a method or apparatus for communicating data between a device and a server.

[0662] "Means for analyzing" refers to a device or method for identifying an object in video content based on the transmitted information and extracting its attribute information.

[0663] "Means for identifying" refers to a device or method that identifies the relevant item from the analyzed data.

[0664] "Attribute information" refers to characteristic information about the identified item (e.g., color, shape, brand logo, etc.).

[0665] The term "means for searching a database" refers to a device or method for obtaining related product information from a database based on the attribute information of a specified item.

[0666] "Product Information" refers to detailed information about related products (e.g., price, image URL, purchase link, etc.).

[0667] "Means for transmitting and displaying to the viewer" refers to a method or apparatus for transmitting product information obtained from the server to the viewer's device and displaying it on the viewer's screen.

[0668] "Past browsing history" and "purchase history" refer to historical information about content that a viewer has viewed or products that they have purchased in the past.

[0669] The term "ranking means" refers to a device or method that evaluates and prioritizes relevant product information based on the viewer's past browsing history and purchasing history.

[0670] The "means for accepting interaction" refers to a method or device that allows a viewer to perform an operation to view detailed information from the provided product information.

[0671] "Means for transitioning to an e-commerce site" refers to a device or method that provides a link to an e-commerce site and directs the viewer to that site for more information.

[0672] The present invention provides a system that allows viewers to easily obtain information about items they are interested in while watching video content on their device, leading to purchasing behavior. This system starts when the viewer taps on an item they are interested in on the screen of their device, and includes a process in which a server recognizes the item and provides related product information.

[0673] Server-side processing

[0674] The server receives video content sent from the viewer's device and analyzes the video frame by frame using an object recognition AI model (e.g., YOLOv5 or Mask R-CNN). The server identifies the items present in each frame and extracts their attributes (e.g., color, shape, brand logo).

[0675] Based on the attributes of the identified objects, an SQL query is executed to search the e-commerce site's database to retrieve matching related products. The server also uses machine learning models (e.g., collaborative filtering and deep learning) to rank the best products based on the viewer's past browsing and purchasing history.

[0676] The server constructs the identified product information and ranking results in JSON format and sends them to the terminal using a RESTful API.

[0677] Terminal side processing

[0678] The device uses a touchscreen event listener to monitor events when a specific item in the video is tapped. The touch location information is captured and temporarily saved.

[0679] This location information and the current frame data are sent to the server using an HTTP POST request. The product information received from the server is analyzed and relevant information is displayed on the screen while the user is watching. Product information includes thumbnail images, prices, purchase links, etc.

[0680] User processing

[0681] Viewers tap on an item of interest (for example, clothing worn by an actress in a drama). The location of the tap is recorded on the device and sent to the server. The viewer checks the product information sent from the server and clicks on a link to the related e-commerce site to view more information. If they find a product they like, they can complete the purchase process through the e-commerce site.

[0682] Specific examples

[0683] For example, a user might be watching a TV drama on their smartphone and become interested in the dress an actress is wearing. When the user taps on the dress, the device detects the tap event and sends the tap location and frame information to the server. The server then analyzes the items in the frame and recognizes the target dress. Based on the recognition results, the database of the e-commerce site is searched to identify the most suitable dress.

[0684] The server sends detailed information about the dress and a link to purchase it to the device, which then displays it to the viewer. The user checks the information displayed, and if they like the dress, they click the provided link to go to the e-commerce site and complete the purchase process.

[0685] This system allows viewers to quickly and easily obtain information about items that interest them, thereby promoting purchasing behavior.

[0686] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0687] Step 1:

[0688] Monitor tap events (device)

[0689] The device monitors events when the viewer taps on a specific item in the video, and uses a touchscreen event listener to obtain the location information when the viewer taps on the item.

[0690] Input: Viewer taps.

[0691] Data processing: Obtain the coordinates of the tapped location and temporarily save them.

[0692] Output: Coordinate information of the tap position.

[0693] Specific behavior:

[0694] Set a touch event listener on the video player.

[0695] Get the tapped coordinate position and save it in a variable.

[0696] Step 2:

[0697] Send tap information to server (device)

[0698] The device sends the acquired tap position information and the current frame data to the server using an HTTP POST request.

[0699] Input: Tap location and frame data.

[0700] Data processing: Convert tap position information and frame data into JSON format.

[0701] Output: The HTTP POST request sent to the server.

[0702] Specific behavior:

[0703] Build tap position information and frame data in JSON format.

[0704] Creates an HTTP POST request and sends it to the server.

[0705] Step 3:

[0706] Video analysis (server)

[0707] The server analyzes the objects in the frame based on the received tap information and frame data, and uses an object recognition AI model (e.g., YOLOv5 or Mask R-CNN) to identify the items and extract their attribute information.

[0708] Input: Tap position information and frame data.

[0709] Data processing: Input the frame data into an object recognition AI model to extract specific item attributes (color, shape, brand logo, etc.).

[0710] Output: Attribute information of the recognized item.

[0711] Specific behavior:

[0712] The frame data stored on the server is input into the object recognition AI model.

[0713] Temporarily save the attribute information of the item obtained from the model.

[0714] Step 4:

[0715] Database search (server)

[0716] The server searches the e-commerce site's database based on the attribute information of the identified item to retrieve related product information. It executes an SQL query to retrieve matching product information from the database.

[0717] Input: Item attribute information.

[0718] Data processing: Generate SQL queries and retrieve product information from the database.

[0719] Output: Retrieved product information.

[0720] Specific behavior:

[0721] Generates an SQL query using the item's attribute information.

[0722] Query the database to retrieve related product information.

[0723] The acquired product information is temporarily saved.

[0724] Step 5:

[0725] Product information ranking (server)

[0726] The server ranks the product information obtained based on the viewer's past browsing and purchasing history, and uses machine learning models (e.g., collaborative filtering and deep learning) to select the best products.

[0727] Input: Retrieved product information, viewer's past browsing and purchasing history.

[0728] Data processing: Input data into a machine learning model and rank it.

[0729] Output: Ranked product information.

[0730] Specific behavior:

[0731] Obtain past viewer data from the database.

[0732] Product information and historical data are input into a machine learning model to obtain ranking results.

[0733] Temporarily save the ranking results.

[0734] Step 6:

[0735] Product information sent to server (server)

[0736] The server sends the ranked product information to the terminal and returns the product information in JSON format using a RESTful API.

[0737] Input: Ranked product information.

[0738] Data processing: Build product information in JSON format.

[0739] Output: The HTTP response sent to the device.

[0740] Specific behavior:

[0741] Build product information in JSON format.

[0742] Use the RESTful API to generate an HTTP response and send it to the device.

[0743] Step 7:

[0744] Display product information (device)

[0745] The terminal analyzes the product information received from the server and displays related information on the screen being viewed.

[0746] Input: The HTTP response from the server.

[0747] Data processing: Analyze JSON data and extract product information.

[0748] Output: Product information displayed on the screen.

[0749] Specific behavior:

[0750] Parse the JSON data received from the server.

[0751] Display the extracted product information on the screen (e.g., by placing a thumbnail image or purchase link).

[0752] Step 8:

[0753] Check detailed information (user)

[0754] The user checks the product information displayed on the terminal and clicks on the link to view more detailed information.

[0755] Input: View product information.

[0756] Data processing: Clicking on a link.

[0757] Output: Show detailed information.

[0758] Specific behavior:

[0759] Check the detailed information of the displayed product.

[0760] Click on the link to access the e-commerce site, if desired.

[0761] Step 9:

[0762] Purchasing decision (user)

[0763] The user decides on a purchase based on the information provided, and if they find a product they like, they complete the purchase process through the e-commerce site.

[0764] Input: Display product details.

[0765] Data processing: Operations to carry out purchase procedures.

[0766] Output: Product purchase completed.

[0767] Specific behavior:

[0768] Check the product details and decide to purchase.

[0769] Proceed through a purchase on an e-commerce site.

[0770] (Application example 1)

[0771] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0772] In today's content distribution services, it is difficult for viewers to quickly and accurately obtain information about items they are interested in while watching a video, leading to immediate purchase behavior. There is also a need for a system that allows viewers to obtain detailed information about items along with related products. Given this situation, a means is needed to efficiently satisfy viewers' interests and encourage purchasing behavior without compromising the viewing experience.

[0773] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0774] In this invention, the server includes means for tapping on an item of interest to a viewer while watching video content on a device, means for transmitting tapped position information and frame data to the server, means for analyzing each video frame using an object recognition AI model to identify the relevant item, means for searching a database based on feature information of the identified item to obtain related product information, means for transmitting the obtained product information to the device and displaying it to the viewer, and means for generating detailed data about the tapped item as a prompt sentence using a generative AI model and presenting it to the viewer. This makes it possible to quickly obtain information about the item being viewed and promote purchasing behavior by providing necessary detailed information while maintaining the viewing experience.

[0775] A "viewer" is a user who watches video content.

[0776] A "device" is a hardware device for viewing video content, and specifically includes smartphones, tablets, and computers.

[0777] "Video content" refers to content that combines video and audio that is viewed by a viewer through a device.

[0778] An "item" is an object or element that appears in video content and is an object that may interest viewers.

[0779] A "tap" is an action in which the viewer touches the device's touch panel and identifies the location.

[0780] "Location information" refers to the coordinate data on the screen of the device that was tapped.

[0781] "Frame data" is still image data that represents a specific moment in video content.

[0782] A "server" is a computer system that receives, processes, and analyzes data sent from devices.

[0783] An "object recognition AI model" is an artificial intelligence algorithm for identifying specific items within a frame of video content.

[0784] "Feature information" is data relating to the attributes and identifiers of the identified item.

[0785] A "database" is a collection of information in which related product information is stored.

[0786] "Obtained product information" is detailed data about products related to the identified item.

[0787] A "generative AI model" is an artificial intelligence algorithm that generates detailed data about a specific item.

[0788] A "prompt" is a detailed explanatory text generated by a generative AI model to be presented to the viewer.

[0789] "Viewing history" is a record of video content that a viewer has viewed in the past.

[0790] "Purchase history" is a record of products that a viewer has purchased in the past.

[0791] "Ranking" is an evaluation process that ranks related products based on viewing history and purchasing history.

[0792] "Interaction" refers to the operations or inputs that viewers make through a device.

[0793] An "EC site" is a website for conducting electronic commerce.

[0794] The present invention provides a system for promptly and accurately providing information about items of interest to viewers while they are watching video content on their devices, thereby promoting purchasing behavior. An embodiment of this system will be described below.

[0795] System configuration

[0796] This system mainly consists of a server, viewer terminals, and a database.

[0797] 1. Server

[0798] Hardware: General cloud server (e.g. AWS EC2)

[0799] Software: Flask (web framework), OpenCV (image processing library), TensorFlow or PyTorch (object recognition AI model), generative AI model

[0800] Function: The server receives tap events from viewers, analyzes video frames to identify items, retrieves relevant product information from a database, and generates detailed prompts using a generative AI model.

[0801] 2. Terminal

[0802] Hardware: Smartphones, tablets

[0803] Software: Tap event monitoring application, HTTP request library

[0804] Function: The device detects tap events while the viewer is watching video content, sends the location information and frame data to the server, and displays the retrieved product information.

[0805] 3. Database

[0806] Components: A database containing product information for an e-commerce site

[0807] Function: Provides related product information based on the characteristics of the identified item.

[0808] Operation flow

[0809] 1. Viewer Controls

[0810] When a viewer is watching video content on a device, they tap on an item they are interested in. At this time, the device transmits information about the tap position and the corresponding frame data to the server.

[0811] 2. Processing on the server

[0812] Based on the received data, the server uses an object recognition AI model to analyze the items in the frame and identify the relevant item. It then searches a database based on the identified item's characteristics (e.g., color, shape, brand) to retrieve related product information. It then uses a generative AI model to generate a detailed prompt for the relevant product and present it to the viewer. This prompt includes the product name, price, and a purchase link.

[0813] 3. Display on device

[0814] The product information received from the server is displayed on the device, and the viewer can check the provided information. If the viewer is interested, they can be redirected to an e-commerce site for more detailed information or to complete the purchase process.

[0815] Specific examples

[0816] While watching a drama on a smartphone, a viewer becomes interested in the blue dress worn by the actress and taps on that part. At the same time, the device sends its location information and frame data to a server. The server uses an object recognition AI model to analyze the items in the frame and identify the "blue dress." It then searches a database to retrieve related product information (product name: "Blue Dress," price: "10,000 yen," purchase link, thumbnail image). It then uses a generative AI model to generate the following prompt:

[0817] "While watching the video, I noticed a blue dress the actress was wearing. When I tapped on the item, more information was displayed on the screen."

[0818] Viewers can check the product information on their smartphones based on the information sent from the server, and then click on the link to be taken to the e-commerce site and complete the purchase process.

[0819] The above is the "Mode for Carrying Out the Invention" of the present invention. This system allows viewers to quickly and accurately obtain information about items that interest them, and promotes purchasing behavior.

[0820] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0821] Step 1:

[0822] Viewers tap on items that interest them while watching video content on their device.

[0823] Input: Viewer taps

[0824] Output: Tap location and frame data of the video being watched

[0825] How it works: The device captures and records the location of the tap and the video frame data at that moment. This information is used to make requests to the server.

[0826] Step 2:

[0827] The tapped position information and frame data are sent to the server.

[0828] Input: Tap location and frame data of the video being watched

[0829] Output: HTTP request to the server

[0830] Operation: The device generates and sends an HTTP request to the server to send the collected location information and frame data.

[0831] Step 3:

[0832] The server analyzes the objects in the frame based on the information received and identifies the relevant item.

[0833] Input: Tap location and frame data

[0834] Output: Feature information of the identified item (e.g. color, shape, brand)

[0835] How it works: The server uses an object recognition AI model to analyze the frame data and identify the item of interest based on the transmitted tap location. From this analysis, it extracts feature information about the item.

[0836] Step 4:

[0837] A database is searched based on the characteristic information of the identified item to obtain related product information.

[0838] Input: Feature information of the identified item

[0839] Output: Related product information (product name, price, purchase link, thumbnail image, etc.)

[0840] Operation: The server uses the acquired feature information to search the database to identify and retrieve relevant product information. The search results include the most relevant product information.

[0841] Step 5:

[0842] Based on the generated product information, a prompt sentence is generated using a generative AI model and presented to the viewer.

[0843] Input: Related product information

[0844] Output: Prompt (e.g., "While watching the video, I noticed the blue dress the actress was wearing. If you tap on the item, more information will appear on the screen.")

[0845] How it works: The server uses a generative AI model to convert the acquired product information into detailed prompts that are used to communicate product details to the viewer.

[0846] Step 6:

[0847] The server sends product information and a prompt to the device.

[0848] Input: Prompt statement and product information

[0849] Output: Sending data to the viewer's device

[0850] How it works: The server generates a response to send the generated prompt and product information to the viewer's device, and sends it to the device using the HTTP protocol.

[0851] Step 7:

[0852] The terminal displays the received product information and prompt text on the screen.

[0853] Input: Received product information and prompt text

[0854] Output: Product information and prompt text displayed on the screen

[0855] Operation: The terminal analyzes the product information and prompt text received from the server and displays it on the screen in a format that is easy for the viewer to see.

[0856] Step 8:

[0857] Viewers click on the provided link to view more information.

[0858] Input: Product information link

[0859] Output: Transition to EC site and display of detailed information

[0860] How it works: When a viewer clicks on a product link displayed on the screen, the device's browser will redirect to the corresponding e-commerce site, where they can view detailed information and complete the purchase process.

[0861] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0862] The present invention is a system that easily obtains information about items that interest a viewer while they are watching video content on their device, and by combining this with an emotion engine, provides information and promotes purchasing behavior that is tailored to the viewer's emotions. This system starts when the viewer taps on an item of interest on the device screen, and includes a process in which a server recognizes the item and provides related product information. The system also uses the emotion engine to analyze the viewer's emotions and optimize the information provided.

[0863] Server-side processing

[0864] 1. Running the video analysis module:

[0865] The server receives the video content and begins analyzing it frame by frame, using an object recognition AI model to identify items within the frame.

[0866] 2. Database search of identification results:

[0867] Based on the information of the recognized object (for example, color, shape, brand logo, etc. for clothing), the e-commerce site's database is searched to identify related product information.

[0868] 3. Ranking of specific products:

[0869] Based on the search results, the AI ​​ranks the most suitable products based on the viewer's past browsing and purchasing history.

[0870] 4. Running the Emotion Engine:

[0871] The server analyzes the viewer's facial expressions and voice data using an emotion engine to identify the viewer's emotional state.

[0872] 5. Emotion-based filtering:

[0873] Filter or rank product information based on the viewer's emotional state.

[0874] 6. Provision of Information:

[0875] Information about the tapped item is instantly sent to the device, and data is prepared that links to the details page of the product the viewer needs.

[0876] Terminal side processing

[0877] 1. Listening for tap events:

[0878] The device monitors the viewer's tap events and sends the location information to the server when a specific item is tapped.

[0879] 2. Sending a request to the server:

[0880] A request is sent to the server along with the frame data of the tap position, and this information is used by the server's analysis module.

[0881] 3. Information Receipt and Display:

[0882] It receives product information sent from the server and displays related information on the screen while the user is viewing the content. It improves user convenience by displaying links to related e-commerce sites and thumbnail images of the products.

[0883] User processing

[0884] 1. Item tap:

[0885] The viewer taps on an item that interests them (e.g., an actress's clothing). This action records the location where the tap was made on the device.

[0886] 2. Information verification:

[0887] Check the product information displayed on your device. Click on the link to the related e-commerce site to view more information.

[0888] 3. Purchasing decision:

[0889] They decide on their purchasing behavior based on the information provided, and if they like the product, they complete the purchase process through the e-commerce site.

[0890] Specific examples

[0891] 1. User Scenario:

[0892] While watching a drama starring an actress on his smartphone, a user becomes interested in the dress the actress is wearing and taps on the dress.

[0893] 2. Tap event occurs:

[0894] The device detects the tap event and sends the tap position and frame information to the server.

[0895] 3. Parsing on the server:

[0896] The server analyzes the items in the frame and recognizes the dress. Based on the recognition results, it searches the e-commerce site database to identify the most suitable dress.

[0897] 4. Running the Emotion Engine:

[0898] The server analyzes the user's emotions from facial expressions and voice data and determines, for example, whether the user is excited.

[0899] 5. Emotional information:

[0900] Based on the information from the emotion engine, the server provides preferentially information on products that suit the user's state of excitement.

[0901] 6. Information provision:

[0902] The server sends the details of the dress and a purchase link to the device, which then displays the details of the dress to the user based on the received information.

[0903] 7. Purchasing behavior:

[0904] The user checks the displayed information, and if they like the dress, they click the provided link to go to the e-commerce site and complete the purchase process.

[0905] According to the above-described embodiment, the present invention can quickly and easily obtain information about items that interest a viewer, and can further promote purchasing behavior by displaying information according to the viewer's emotions.

[0906] The processing flow will be explained below.

[0907] Step 1:

[0908] A user watches video content on a device and taps an item that interests them (e.g., an actress's clothing). This action records the location of the tap on the device.

[0909] Step 2:

[0910] The device detects a touch event and obtains the touch location, the current video frame data, and a timestamp, including the time at which the operation occurred.

[0911] Step 3:

[0912] The device sends the tap location, frame data, and timestamp to the server in a request that also includes the necessary metadata.

[0913] Step 4:

[0914] Based on the tap location information and frame data received by the server, a video analysis module is used to analyze the objects in the frame, using an object recognition AI model to identify the item closest to the tapped location.

[0915] Step 5:

[0916] The server extracts the characteristic information (e.g., color, shape, brand, pattern) of the identified item and searches the e-commerce site's database to obtain related product information.

[0917] Step 6:

[0918] The server retrieves relevant product information from the search results and ranks the most suitable products based on the viewer's past browsing and purchasing history, taking user preferences into account in the process.

[0919] Step 7:

[0920] The server runs an emotion engine that analyzes the viewer's facial and voice data to determine their emotions, in this case emotional states such as excitement, joy, and surprise.

[0921] Step 8:

[0922] The server further filters the ranked product information based on the user's emotional state, prioritizing products that best suit the user's current emotions.

[0923] Step 9:

[0924] The server sends the identified product information (product name, price, sales site link, image, etc.) to the device. This information is optimized based on emotions.

[0925] Step 10:

[0926] The product information received by the device is displayed on the screen in a format that is intuitive and easy for viewers to understand, including pop-ups and overlays.

[0927] Step 11:

[0928] Users review the displayed product information and click on the link to the product detail page of the product they are interested in. This action makes it easier to obtain additional information.

[0929] Step 12:

[0930] The device will redirect to an e-commerce site where the user can view detailed information, allowing them to make a final purchasing decision.

[0931] Example 2

[0932] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0933] Conventional video viewing systems lack a means for viewers to quickly and easily obtain information about items that interest them within video content. Furthermore, there is no method for providing information or encouraging purchasing behavior based on the viewer's emotional state. This makes it difficult to effectively stimulate viewers' purchasing desire and provide an optimal purchasing experience.

[0934] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for tapping an item in which a viewer is interested, a means for transmitting tapped position information and frame data to the server, a means for analyzing an object in the frame based on the transmitted information and identifying the relevant item, a means for searching a database based on characteristic information of the identified item and acquiring related product information, a means for analyzing facial expressions and voice data of the viewer and identifying the emotional state, and a means for filtering or ranking product information based on the identified emotional state. This makes it possible to quickly and easily acquire information about items in which the viewer is interested, and to optimally display information and promote purchases according to the viewer's emotions.

[0935] A "viewer" is a user who is viewing video content on a device.

[0936] A "device" is an electronic device (e.g., smartphone, tablet, computer) that displays video content and can be operated by a user.

[0937] "Video content" means digital media containing a combination of moving images and sounds that convey information through visual and auditory means.

[0938] An "item" refers to an object or person that appears in video content and is an object of interest to viewers.

[0939] "Tap" means that the viewer selects an item by touching the device's touchscreen.

[0940] "Tap location information" is the coordinate data when a viewer taps an item on the device screen.

[0941] "Frame data" refers to information about each frame (still image) of video content, and is data that represents video at a specific point in time.

[0942] A "server" is a central processing unit that receives data from devices and performs analytical processing and database searches.

[0943] "Object analysis means" refers to the technology or algorithms used by the server to identify objects or items present in each frame of video content (e.g., object recognition AI models).

[0944] The "means for identifying the relevant item" is a method for finding the item that the viewer tapped on from among the analyzed objects.

[0945] The "means for searching a database" is a method for querying a database in which various product information is stored based on the characteristic information of the identified item.

[0946] "Related Product Information" is detailed information (e.g., price, description, images, links) about products that are similar or match the identified item.

[0947] The "means for displaying to the viewer" refers to a method for visually presenting the acquired product information on a device.

[0948] "Means for analyzing facial and / or audio data" means the techniques or algorithms used to infer the emotional state of a viewer from their facial expressions or audio (e.g., emotion engine).

[0949] "Emotional state" is data that reflects the type and intensity of a viewer's emotion, including joy, excitement, sadness, etc.

[0950] A "filtering or ranking means" is a method for sorting or prioritizing retrieved product information based on the identified emotional state.

[0951] The present invention is a system that allows a viewer to quickly and easily obtain information about items that interest them while watching video content on a device, and provides information that corresponds to the viewer's emotions and promotes purchasing behavior.

[0952] The overall operation of the system involves three entities: the server, the device, and the viewer. We will explain how each entity interacts and processes information.

[0953] Server-side processing

[0954] First, we will explain the server-side processing. Specifically, it has the following functions:

[0955] 1. Video Analysis Module:

[0956] The server splits the video content received from the user into frames and identifies items within the frames using an object recognition AI model (e.g., YOLO or Detectron2). Through this analysis, the server identifies items that may be of interest to the viewer.

[0957] 2. Database Search:

[0958] Based on the analyzed item's attribute information (color, shape, brand logo, etc.), the server queries the e-commerce site's database (e.g., Amazon Product Advertising API, Rakuten product database) to obtain related product information.

[0959] 3. Ranking Algorithm:

[0960] Based on the acquired product information, the server takes into account the viewer's past browsing and purchasing history and uses AI algorithms (such as collaborative filtering or content-based filtering) to rank the products that are best suited to the viewer.

[0961] 4. Sentiment Analysis Module:

[0962] The server acquires the viewer's facial expression and voice data and analyzes it using an emotion engine (e.g., Microsoft Azure Emotion API, IBM Watson Tone Analyzer) to determine their real-time emotional state (excitement, joy, calmness, etc.).

[0963] 5. Emotion-based filtering:

[0964] The server filters or re-ranks product information based on the viewer's emotional state, for example, prioritizing special offers and limited edition products for an excited viewer.

[0965] 6. Provision of Information:

[0966] The server immediately transmits the filtered product information to the terminal and provides a link to the detailed page of the product the viewer needs.

[0967] Terminal side processing

[0968] Next, the processing on the terminal side will be described.

[0969] 1. Listening for tap events:

[0970] The device monitors the viewer's tapping on the screen while watching a video, acquires the location information of the tap, and sends this location information to the server.

[0971] 2. Data transmission:

[0972] The tapped position information and frame data are sent to the server, which uses this information in the server's video analysis module.

[0973] 3. Information Receipt and Display:

[0974] It receives product information sent from the server and displays related information on the screen while the user is watching. Specifically, it displays product information pop-ups, links to related e-commerce sites, thumbnail images of the products, and more.

[0975] User processing

[0976] Finally, the processing on the user side will be described.

[0977] 1. Item tap:

[0978] The user taps on an item of interest (e.g., the actress's clothing) in the video they are watching. This action records the location of the tap on the device.

[0979] 2. Information verification:

[0980] The user checks the product information displayed on the device, clicks on the link to the related e-commerce site, and views the product details page.

[0981] 3. Purchasing decision:

[0982] The user decides on a purchasing decision based on the information provided, and if they like the product, they complete the purchase process through the e-commerce site.

[0983] Specific examples

[0984] Next, a specific example of the operation of the present invention will be described.

[0985] 1. User Scenario:

[0986] While watching a drama starring an actress on his smartphone, a user becomes interested in the dress the actress is wearing and taps on the dress.

[0987] 2. Tap event occurs:

[0988] The device detects the tap event and sends the tap position and frame information to the server. Specifically, it sends the tap coordinates and the frame data at that time.

[0989] 3. Parsing on the server:

[0990] The server analyzes the items in the received frame and recognizes the dress. Based on the recognition results, it searches the e-commerce site database to identify the most suitable dress.

[0991] 4. Running the Emotion Engine:

[0992] The server analyzes the user's emotions from facial expressions and voice data, and determines, for example, whether the user is in an excited state.

[0993] 5. Emotional information:

[0994] The server provides information on products that suit the user's state of excitement with priority based on information from the emotion engine.

[0995] 6. Information provision:

[0996] The server sends detailed information about the dress and a link to purchase it to the device, which then displays the details of the dress to the user based on the information received.

[0997] 7. Purchasing behavior:

[0998] The user checks the displayed information, and if they like the dress, they click on the provided link to go to the e-commerce site and complete the purchase process.

[0999] Example prompt:

[1000] "If I want to know about the dress an actress is wearing, how can the server provide this information and explain the process for the user to purchase it?"

[1001] As described above, the system of the present invention is designed to quickly and easily obtain information about items that interest viewers, and to display information optimally in response to the viewers' emotions and promote purchases.

[1002] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1003] Step 1:

[1004] Listening for touch events

[1005] The device monitors the viewer's tapping on the screen while watching a video. The input is the viewer's tapping on the touchscreen. Specifically, the device obtains the tapped position information (coordinate data) and the frame data indicating which frame was being displayed. The output is the tapped position information and frame data.

[1006] Step 2:

[1007] Sending a request to the server

[1008] The device sends the acquired tap position information and frame data to the server. The inputs include tap position information and frame data. Specifically, the device sends this information to the server in the form of an HTTP request or similar. The output includes the tap position information and frame data sent to the server.

[1009] Step 3:

[1010] Running the video analysis module

[1011] The server uses the received frame data to run an object recognition AI model (e.g., YOLO or Detectron2) and analyze the objects in the frame. The input is the frame data. Specifically, the server analyzes the frame and identifies the item. The output is information about the analyzed item.

[1012] Step 4:

[1013] Database search of identification results

[1014] The server searches the e-commerce site's database based on the analyzed item's characteristic information (color, shape, brand logo, etc.). The analyzed item information is the input. Specifically, the server calls the API to retrieve related product information. The retrieved product information is the output.

[1015] Step 5:

[1016] Ranking of specific products

[1017] The server uses AI algorithms (such as collaborative filtering or content-based filtering) to rank the acquired product information. The inputs are the acquired product information and the viewer's past browsing history and purchase history. Specifically, the server uses AI to rank the products. The output is the ranked product information.

[1018] Step 6:

[1019] Running the Emotion Engine

[1020] The server acquires the viewer's facial expression and voice data and analyzes it using an emotion engine (e.g., Microsoft Azure Emotion API, IBM Watson Tone Analyzer). The input is the viewer's facial expression and voice data. Specifically, the server analyzes the emotion data and identifies the viewer's emotional state. The output is the analyzed emotional state.

[1021] Step 7:

[1022] Sentiment-based filtering

[1023] The server filters or re-ranks the product information based on the analyzed emotional state. The inputs include the analyzed emotional state and the ranked product information. As a specific operation, the server sorts or prioritizes the product information according to the emotional state. The output includes the filtered or re-ranked product information.

[1024] Step 8:

[1025] Providing information

[1026] The server transmits the filtered or re-ranked product information to the terminal. The input includes the filtered or re-ranked product information. As a specific operation, the server converts the product information into a data packet and transmits it to the terminal. The output includes the product information transmitted to the terminal.

[1027] Step 9:

[1028] Information reception and display

[1029] The device receives product information sent from the server and displays related information on the screen being viewed. The input is the product information sent from the server. In concrete terms, the device presents the product information to the user in the form of a pop-up display or related links. The output is the product information displayed in a format that the viewer can check.

[1030] Step 10:

[1031] Information verification and purchasing behavior

[1032] The user checks the product information displayed on the device and, if interested, proceeds with the purchase. The input is the product information displayed on the device. The specific operation is for the user to click on a link to the relevant e-commerce site and proceed with the purchase. The output is a result indicating whether the purchase was completed.

[1033] Through the above steps, viewers can easily obtain information about items that interest them, and information is displayed according to their emotions, promoting purchasing behavior.

[1034] (Application example 2)

[1035] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1036] Currently, there are insufficient ways for viewers to easily obtain information about items they are interested in while watching video content.In addition, information is not provided according to the viewer's emotional state, making it difficult to suggest optimal product information.

[1037] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1038] In this invention, the server includes means for tapping on an object of interest to a viewer while watching video content on a device, means for transmitting tapped position information and frame data to the server, means for analyzing objects in the frame based on the transmitted information and identifying the relevant object, means for searching a database based on characteristic information of the identified object and acquiring related product information, means for transmitting the acquired product information to the device and displaying it to the viewer, means for analyzing the viewer's facial expressions and voice data using an emotion engine and identifying the viewer's emotional state, and means for filtering or ranking the product information to be provided based on the viewer's emotional state. This allows the viewer to quickly and easily obtain information on items that interest them while watching a video, and further makes it possible to provide optimal product information according to the viewer's emotional state.

[1039] A "viewer" is a person who watches video content.

[1040] A "device" is an electronic device that displays video content and allows viewers to interact with it.

[1041] "Video content" refers to all media content provided in video format.

[1042] An "object" is an object that a viewer becomes interested in while watching video content and tries to obtain information by tapping on it.

[1043] A "tap" is the act of touching a specific location on a device screen once with a finger or stylus.

[1044] "Frame data" refers to each still image data that constitutes video content.

[1045] A "server" is a computer system that receives a request over a network, executes the specified process, and returns a response.

[1046] "Analyzing an object" means recognizing a specific object from frame data in video content and extracting information about it.

[1047] "Feature information" refers to attribute information for identifying an object, such as the object's color, shape, brand logo, etc.

[1048] A "database" is a collection of digital information organized to allow efficient storage and retrieval of specific information.

[1049] "Product Information" refers to detailed information about products and services related to the subject matter.

[1050] An "emotion engine" is a software module that analyzes the viewer's emotional state and optimizes the information provided based on the results.

[1051] "Emotional state" refers to data that represents the viewer's current psychological and physiological state.

[1052] "Filtering" refers to the act of sorting and prioritizing the information provided.

[1053] "Ranking" is the act of assigning a rank to information or objects based on certain criteria.

[1054] The present invention is a system that easily obtains information about objects of interest to viewers while they are watching video content on their devices, and provides optimal product information according to the viewer's emotional state. The system for implementing the present invention consists of a server, a terminal, and user interactions.

[1055] Server Processing

[1056] The server consists of the following modules:

[1057] Object Recognition Module: Using TensorFlow and OpenCV, this module analyzes and identifies objects within frames of video content. This module recognizes objects based on the transmitted location information and frame data, and extracts their feature information.

[1058] Database search module: Using MySQL or MongoDB, this module searches the database for relevant product information based on the characteristics of the identified object. This module quickly retrieves and ranks product information.

[1059] Sentiment analysis engine: Using Google Cloud Vision and Amazon Rekognition, the engine analyzes the viewer's emotional state from facial expressions and voice data, and filters or ranks the product information provided based on the analysis results.

[1060] Terminal handling

[1061] The terminal has the following features:

[1062] Monitoring tap events: Events that occur when a viewer taps the screen of the device are monitored, and the location information and frame data are sent to the server.

[1063] Communication with the server: When a tap event is detected, the information is sent to the server as a request and product information sent from the server is received.

[1064] Information display: Product information obtained from the server is displayed on the screen while the content is being viewed, and related product links and thumbnail images are provided.

[1065] User Action

[1066] The user interacts with:

[1067] Item Tap: Tap on an object that interests you while watching and send its location information.

[1068] Check information: Check the product information displayed on your device and click on the link for more information or to purchase.

[1069] Purchasing behavior: Select a product based on the information provided and complete the purchase process on the e-commerce site.

[1070] Specific examples

[1071] For example, suppose a user is watching a movie on a media distribution platform and is interested in the clothing worn by the lead actor. In this case, the user taps on the device screen. This tap event is detected by the device, and the relevant location information and frame data are sent to the server. The server identifies the clothing based on this data and searches for related product information from a database based on its feature information. Furthermore, an emotion analysis engine analyzes the user's emotional state (e.g., excitement or curiosity), ranks the product information accordingly, and sends it to the device. The device displays the results to the user, who can click on the provided link to check more information and make a purchase.

[1072] Prompt Sentence Examples

[1073] While a user is watching a movie on their smartphone, they tap on an item they are interested in. Can you give us an example of a system that retrieves information about that item and provides relevant product information, taking into account the user's emotional state?

[1074] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1075] Step 1:

[1076] User interaction and tap events

[1077] While a user is watching video content on a device, they tap on an object that interests them.

[1078] Input: User tap position information and video frame data

[1079] What happens: The device's touchscreen detects the user's taps and captures their location.

[1080] Step 2:

[1081] Sending tap position

[1082] The terminal transmits the tap position information and the associated frame data to the server.

[1083] Input: User tap position information and video frame data

[1084] Output: Tap position information and frame data are sent to the server.

[1085] Specific operation: The device detects a touch event and sends the coordinates and current frame data to the server as an HTTP request.

[1086] Step 3:

[1087] Analysis of objects in the frame

[1088] The server analyzes the objects in the frame based on the transmitted location information and frame data, and identifies the relevant target.

[1089] Input: Tap position information and video frame data

[1090] Output: Feature information of identified objects

[1091] Specific operation: The object recognition module (TensorFlow or OpenCV) on the server recognizes the object around the tap position in the frame and extracts its feature information.

[1092] Step 4:

[1093] Database search

[1094] The server searches the database based on the characteristic information of the identified object and obtains related product information.

[1095] Input: Feature information of the identified object

[1096] Output: List of product details

[1097] Specific operation: Using the feature information, query the database (MySQL or MongoDB) to retrieve the corresponding product information.

[1098] Step 5:

[1099] Emotional state analysis

[1100] The server uses an emotion analysis engine to identify the viewer's emotional state based on the user's facial expression data and voice data.

[1101] Input: User's facial expression data and voice data

[1102] Output: Emotional state assessment result

[1103] Specific operation: Analyzes the user's emotional state using APIs such as Google Cloud Vision and Amazon Rekognition.

[1104] Step 6:

[1105] Filtering and ranking product information

[1106] The server filters or ranks the product information based on the analyzed emotional state.

[1107] Input: List of product information, emotional state evaluation results

[1108] Output: Filtered or ranked product information

[1109] Specific behavior: Select and prioritize the most relevant product information based on emotional state.

[1110] Step 7:

[1111] Sending and Displaying Information

[1112] The server transmits the filtered product information to the terminal, which displays it on the screen being viewed.

[1113] Input: Filtered or ranked product information

[1114] Output: Display information on the user's device (product image, link to details, etc.)

[1115] Specific operation: The server sends the generated product information back to the device as an HTTP response, and the device receives the information and displays it on the screen.

[1116] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1117] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1118] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1119] [Third embodiment]

[1120] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1121] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1122] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1123] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1124] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1125] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1126] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1127] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1128] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1129] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1130] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1131] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1132] The present invention provides a system that allows viewers to easily obtain information about items they are interested in while watching video content on their device, leading to purchasing behavior. This system starts when the viewer taps on an item they are interested in on the device screen, and includes a process in which a server recognizes the item and provides related product information.

[1133] Server-side processing

[1134] 1. Running the video analysis module:

[1135] The server receives the video content and begins analyzing it frame by frame, using an object recognition AI model to identify items within the frame.

[1136] 2. Database search of identification results:

[1137] Based on the information of the recognized object (for example, color, shape, brand logo, etc. for clothing), the e-commerce site's database is searched to identify related product information.

[1138] 3. Ranking of specific products:

[1139] Based on the search results, the AI ​​ranks the most suitable products based on the viewer's past browsing and purchasing history.

[1140] 4. Provision of Information:

[1141] Information about the tapped item is instantly sent to the device, and data is prepared that links to the details page of the product the viewer needs.

[1142] Terminal side processing

[1143] 1. Listening for tap events:

[1144] The device monitors the viewer's tap events and sends the location information to the server when a specific item is tapped.

[1145] 2. Sending a request to the server:

[1146] A request is sent to the server along with the frame data of the tap position, and this information is used by the server's analysis module.

[1147] 3. Information Receipt and Display:

[1148] It receives product information sent from the server and displays related information on the screen while the user is viewing the content. It improves user convenience by displaying links to related e-commerce sites and thumbnail images of the products.

[1149] User processing

[1150] 1. Item tap:

[1151] The viewer taps on an item that interests them (e.g., the clothes worn by an actress). This action records the location where the tap was made on the device.

[1152] 2. Information verification:

[1153] Check the product information displayed on your device. Click on the link to the related e-commerce site to view more information.

[1154] 3. Purchasing decision:

[1155] They decide on their purchasing behavior based on the information provided, and if they like the product, they complete the purchase process through the e-commerce site.

[1156] Specific examples

[1157] 1. User Scenario:

[1158] While watching a drama starring an actress on his smartphone, a user becomes interested in the dress the actress is wearing and taps on the dress.

[1159] 2. Tap event occurs:

[1160] The device detects the tap event and sends the tap position and frame information to the server.

[1161] 3. Parsing on the server:

[1162] The server analyzes the items in the frame and recognizes the dress. Based on the recognition results, it searches the e-commerce site database to identify the most suitable dress.

[1163] 4. Information provision:

[1164] The server sends the details of the dress and a purchase link to the device, which then displays the details of the dress to the user based on the received information.

[1165] 5. Purchasing behavior:

[1166] The user checks the displayed information, and if they like the dress, they click the provided link to go to the e-commerce site and complete the purchase process.

[1167] In the above-described form, the present invention allows viewers to quickly and easily obtain information about items that interest them, thereby promoting purchasing behavior.

[1168] The processing flow will be explained below.

[1169] Step 1:

[1170] A user watches video content on a device and taps on an item of interest (e.g., an actress's clothing).

[1171] Step 2:

[1172] The device detects the touch event and retrieves the touch location and the current video frame data, including a specific timestamp in the video.

[1173] Step 3:

[1174] The device sends the tap location, frame data, and timestamp to the server, along with any necessary metadata.

[1175] Step 4:

[1176] Based on the tap position information and frame data received by the server, a video analysis module is used to analyze the object in the corresponding frame.

[1177] Step 5:

[1178] The server uses the analysis to identify the item closest to the tapped location, which is performed by an object recognition AI model.

[1179] Step 6:

[1180] The server extracts the characteristic information of the identified item (e.g., color, shape, brand, pattern) and searches the e-commerce site's database.

[1181] Step 7:

[1182] The server retrieves relevant product information from the database and ranks the most suitable products by referring to the viewer's past browsing and purchasing history.

[1183] Step 8:

[1184] The server sends the ranked product information (product name, price, sales site link, image, etc.) to the terminal.

[1185] Step 9:

[1186] The product information received by the device is displayed on the screen in a pop-up or overlay format.

[1187] Step 10:

[1188] The user checks the displayed product information and clicks on the link to the product detail page of the product that interests them.

[1189] Step 11:

[1190] The device will redirect to the EC site, where the user can view detailed information and then complete the purchase procedure.

[1191] Example 1

[1192] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1193] When viewers watch video content on their devices, there is a demand for a system that can quickly and easily obtain information about items that interest them and directly link this information to purchasing behavior. Conventional systems have the problem that the process from finding a product that interests a viewer to purchasing it is complicated and time-consuming. Furthermore, they lack a mechanism for suggesting optimal products based on the viewer's past preferences and browsing history.

[1194] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1195] In this invention, the server includes means for analyzing objects in the frame based on the transmitted information and identifying the relevant item, means for searching a database based on attribute information of the identified item and acquiring related product information, and means for transmitting the acquired product information to the device and displaying it to the viewer. This enables a viewer to quickly acquire product information related to an item that interests them in the video, and receive appropriate product suggestions, simply by tapping on that item.

[1196] A "viewer" is someone who watches video content on a device and interacts with items of interest.

[1197] A "device" is a device that a viewer uses to watch video content, and includes, for example, a smartphone, a tablet, a PC, and the like.

[1198] "Video content" means media, including video and audio, that is played by a viewer on a device.

[1199] "Item" refers to an object or subject that appears within video content.

[1200] "Tapping means" refers to a device or method by which a viewer operates the touchscreen of a device to select a particular item.

[1201] "Tap location information" refers to the coordinate information when a viewer taps an item on the device screen.

[1202] "Frame data" refers to image data of a specific moment (frame) in video content.

[1203] "Server" refers to a computer system that receives and processes data sent from a viewer's device.

[1204] "Means for transmitting" refers to a method or apparatus for communicating data between a device and a server.

[1205] "Means for analyzing" refers to a device or method for identifying an object in video content based on the transmitted information and extracting its attribute information.

[1206] "Means for identifying" refers to a device or method that identifies the relevant item from the analyzed data.

[1207] "Attribute information" refers to characteristic information about the identified item (e.g., color, shape, brand logo, etc.).

[1208] The term "means for searching a database" refers to a device or method for obtaining related product information from a database based on the attribute information of a specified item.

[1209] "Product Information" refers to detailed information about related products (e.g., price, image URL, purchase link, etc.).

[1210] "Means for transmitting and displaying to the viewer" refers to a method or apparatus for transmitting product information obtained from the server to the viewer's device and displaying it on the viewer's screen.

[1211] "Past browsing history" and "purchase history" refer to historical information about content that a viewer has viewed or products that they have purchased in the past.

[1212] The term "ranking means" refers to a device or method that evaluates and prioritizes relevant product information based on the viewer's past browsing history and purchasing history.

[1213] The "means for accepting interaction" refers to a method or device that allows a viewer to perform an operation to view detailed information from the provided product information.

[1214] "Means for transitioning to an e-commerce site" refers to a device or method that provides a link to an e-commerce site and directs the viewer to that site for more information.

[1215] The present invention provides a system that allows viewers to easily obtain information about items they are interested in while watching video content on their device, leading to purchasing behavior. This system starts when the viewer taps on an item they are interested in on the screen of their device, and includes a process in which a server recognizes the item and provides related product information.

[1216] Server-side processing

[1217] The server receives video content sent from the viewer's device and analyzes the video frame by frame using an object recognition AI model (e.g., YOLOv5 or Mask R-CNN). The server identifies the items present in each frame and extracts their attributes (e.g., color, shape, brand logo).

[1218] Based on the attributes of the identified objects, an SQL query is executed to search the e-commerce site's database to retrieve matching related products. The server also uses machine learning models (e.g., collaborative filtering and deep learning) to rank the best products based on the viewer's past browsing and purchasing history.

[1219] The server constructs the identified product information and ranking results in JSON format and sends them to the terminal using a RESTful API.

[1220] Terminal side processing

[1221] The device uses a touchscreen event listener to monitor events when a specific item in the video is tapped. The touch location information is captured and temporarily saved.

[1222] This location information and the current frame data are sent to the server using an HTTP POST request. The product information received from the server is analyzed and relevant information is displayed on the screen while the user is watching. Product information includes thumbnail images, prices, purchase links, etc.

[1223] User processing

[1224] Viewers tap on an item of interest (for example, clothing worn by an actress in a drama). The location of the tap is recorded on the device and sent to the server. The viewer checks the product information sent from the server and clicks on a link to the related e-commerce site to view more information. If they find a product they like, they can complete the purchase process through the e-commerce site.

[1225] Specific examples

[1226] For example, a user might be watching a TV drama on their smartphone and become interested in the dress an actress is wearing. When the user taps on the dress, the device detects the tap event and sends the tap location and frame information to the server. The server then analyzes the items in the frame and recognizes the target dress. Based on the recognition results, the database of the e-commerce site is searched to identify the most suitable dress.

[1227] The server sends detailed information about the dress and a link to purchase it to the device, which then displays it to the viewer. The user checks the information displayed, and if they like the dress, they click the provided link to go to the e-commerce site and complete the purchase process.

[1228] This system allows viewers to quickly and easily obtain information about items that interest them, thereby promoting purchasing behavior.

[1229] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1230] Step 1:

[1231] Monitor tap events (device)

[1232] The device monitors events when the viewer taps on a specific item in the video, and uses a touchscreen event listener to obtain the location information when the viewer taps on the item.

[1233] Input: Viewer taps.

[1234] Data processing: Obtain the coordinates of the tapped location and temporarily save them.

[1235] Output: Coordinate information of the tap position.

[1236] Specific behavior:

[1237] Set a touch event listener on the video player.

[1238] Get the tapped coordinate position and save it in a variable.

[1239] Step 2:

[1240] Send tap information to server (device)

[1241] The device sends the acquired tap position information and the current frame data to the server using an HTTP POST request.

[1242] Input: Tap location and frame data.

[1243] Data processing: Convert tap position information and frame data into JSON format.

[1244] Output: The HTTP POST request sent to the server.

[1245] Specific behavior:

[1246] Build tap position information and frame data in JSON format.

[1247] Creates an HTTP POST request and sends it to the server.

[1248] Step 3:

[1249] Video analysis (server)

[1250] The server analyzes the objects in the frame based on the received tap information and frame data, and uses an object recognition AI model (e.g., YOLOv5 or Mask R-CNN) to identify the items and extract their attribute information.

[1251] Input: Tap position information and frame data.

[1252] Data processing: Input the frame data into an object recognition AI model to extract specific item attributes (color, shape, brand logo, etc.).

[1253] Output: Attribute information of the recognized item.

[1254] Specific behavior:

[1255] The frame data stored on the server is input into the object recognition AI model.

[1256] Temporarily save the attribute information of the item obtained from the model.

[1257] Step 4:

[1258] Database search (server)

[1259] The server searches the e-commerce site's database based on the attribute information of the identified item to retrieve related product information. It executes an SQL query to retrieve matching product information from the database.

[1260] Input: Item attribute information.

[1261] Data processing: Generate SQL queries and retrieve product information from the database.

[1262] Output: Retrieved product information.

[1263] Specific behavior:

[1264] Generates an SQL query using the item's attribute information.

[1265] Query the database to retrieve related product information.

[1266] The acquired product information is temporarily saved.

[1267] Step 5:

[1268] Product information ranking (server)

[1269] The server ranks the product information obtained based on the viewer's past browsing and purchasing history, and uses machine learning models (e.g., collaborative filtering and deep learning) to select the best products.

[1270] Input: Retrieved product information, viewer's past browsing and purchasing history.

[1271] Data processing: Input data into a machine learning model and rank it.

[1272] Output: Ranked product information.

[1273] Specific behavior:

[1274] Obtain past viewer data from the database.

[1275] Product information and historical data are input into a machine learning model to obtain ranking results.

[1276] Temporarily save the ranking results.

[1277] Step 6:

[1278] Product information sent to server (server)

[1279] The server sends the ranked product information to the terminal and returns the product information in JSON format using a RESTful API.

[1280] Input: Ranked product information.

[1281] Data processing: Build product information in JSON format.

[1282] Output: The HTTP response sent to the device.

[1283] Specific behavior:

[1284] Build product information in JSON format.

[1285] Use the RESTful API to generate an HTTP response and send it to the device.

[1286] Step 7:

[1287] Display product information (device)

[1288] The terminal analyzes the product information received from the server and displays related information on the screen being viewed.

[1289] Input: The HTTP response from the server.

[1290] Data processing: Analyze JSON data and extract product information.

[1291] Output: Product information displayed on the screen.

[1292] Specific behavior:

[1293] Parse the JSON data received from the server.

[1294] Display the extracted product information on the screen (e.g., by placing a thumbnail image or purchase link).

[1295] Step 8:

[1296] Check detailed information (user)

[1297] The user checks the product information displayed on the terminal and clicks on the link to view more detailed information.

[1298] Input: View product information.

[1299] Data processing: Clicking on a link.

[1300] Output: Show detailed information.

[1301] Specific behavior:

[1302] Check the detailed information of the displayed product.

[1303] Click on the link to access the e-commerce site, if desired.

[1304] Step 9:

[1305] Purchasing decision (user)

[1306] The user decides on a purchase based on the information provided, and if they find a product they like, they complete the purchase process through the e-commerce site.

[1307] Input: Display product details.

[1308] Data processing: Operations to carry out purchase procedures.

[1309] Output: Product purchase completed.

[1310] Specific behavior:

[1311] Check the product details and decide to purchase.

[1312] Proceed through a purchase on an e-commerce site.

[1313] (Application example 1)

[1314] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1315] In today's content distribution services, it is difficult for viewers to quickly and accurately obtain information about items they are interested in while watching a video, leading to immediate purchase behavior. There is also a need for a system that allows viewers to obtain detailed information about items along with related products. Given this situation, a means is needed to efficiently satisfy viewers' interests and encourage purchasing behavior without compromising the viewing experience.

[1316] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1317] In this invention, the server includes means for tapping on an item of interest to a viewer while watching video content on a device, means for transmitting tapped position information and frame data to the server, means for analyzing each video frame using an object recognition AI model to identify the relevant item, means for searching a database based on feature information of the identified item to obtain related product information, means for transmitting the obtained product information to the device and displaying it to the viewer, and means for generating detailed data about the tapped item as a prompt sentence using a generative AI model and presenting it to the viewer. This makes it possible to quickly obtain information about the item being viewed and promote purchasing behavior by providing necessary detailed information while maintaining the viewing experience.

[1318] A "viewer" is a user who watches video content.

[1319] A "device" is a hardware device for viewing video content, and specifically includes smartphones, tablets, and computers.

[1320] "Video content" refers to content that combines video and audio that is viewed by a viewer through a device.

[1321] An "item" is an object or element that appears in video content and is an object that may interest viewers.

[1322] A "tap" is an action in which the viewer touches the device's touch panel and identifies the location.

[1323] "Location information" refers to the coordinate data on the screen of the device that was tapped.

[1324] "Frame data" is still image data that represents a specific moment in video content.

[1325] A "server" is a computer system that receives, processes, and analyzes data sent from devices.

[1326] An "object recognition AI model" is an artificial intelligence algorithm for identifying specific items within a frame of video content.

[1327] "Feature information" is data relating to the attributes and identifiers of the identified item.

[1328] A "database" is a collection of information in which related product information is stored.

[1329] "Obtained product information" is detailed data about products related to the identified item.

[1330] A "generative AI model" is an artificial intelligence algorithm that generates detailed data about a specific item.

[1331] A "prompt" is a detailed explanatory text generated by a generative AI model to be presented to the viewer.

[1332] "Viewing history" is a record of video content that a viewer has viewed in the past.

[1333] "Purchase history" is a record of products that a viewer has purchased in the past.

[1334] "Ranking" is an evaluation process that ranks related products based on viewing history and purchasing history.

[1335] "Interaction" refers to the operations or inputs that viewers make through a device.

[1336] An "EC site" is a website for conducting electronic commerce.

[1337] The present invention provides a system for promptly and accurately providing information about items of interest to viewers while they are watching video content on their devices, thereby promoting purchasing behavior. An embodiment of this system will be described below.

[1338] System configuration

[1339] This system mainly consists of a server, viewer terminals, and a database.

[1340] 1. Server

[1341] Hardware: General cloud server (e.g. AWS EC2)

[1342] Software: Flask (web framework), OpenCV (image processing library), TensorFlow or PyTorch (object recognition AI model), generative AI model

[1343] Function: The server receives tap events from viewers, analyzes video frames to identify items, retrieves relevant product information from a database, and generates detailed prompts using a generative AI model.

[1344] 2. Terminal

[1345] Hardware: Smartphones, tablets

[1346] Software: Tap event monitoring application, HTTP request library

[1347] Function: The device detects tap events while the viewer is watching video content, sends the location information and frame data to the server, and displays the retrieved product information.

[1348] 3. Database

[1349] Components: A database containing product information for an e-commerce site

[1350] Function: Provides related product information based on the characteristics of the identified item.

[1351] Operation flow

[1352] 1. Viewer Controls

[1353] When a viewer is watching video content on a device, they tap on an item they are interested in. At this time, the device transmits information about the tap position and the corresponding frame data to the server.

[1354] 2. Processing on the server

[1355] Based on the received data, the server uses an object recognition AI model to analyze the items in the frame and identify the relevant item. It then searches a database based on the identified item's characteristics (e.g., color, shape, brand) to retrieve related product information. It then uses a generative AI model to generate a detailed prompt for the relevant product and present it to the viewer. This prompt includes the product name, price, and a purchase link.

[1356] 3. Display on device

[1357] The product information received from the server is displayed on the device, and the viewer can check the provided information. If the viewer is interested, they can be redirected to an e-commerce site for more detailed information or to complete the purchase process.

[1358] Specific examples

[1359] While watching a drama on a smartphone, a viewer becomes interested in the blue dress worn by the actress and taps on that part. At the same time, the device sends its location information and frame data to a server. The server uses an object recognition AI model to analyze the items in the frame and identify the "blue dress." It then searches a database to retrieve related product information (product name: "Blue Dress," price: "10,000 yen," purchase link, thumbnail image). It then uses a generative AI model to generate the following prompt:

[1360] "While watching the video, I noticed a blue dress the actress was wearing. When I tapped on the item, more information was displayed on the screen."

[1361] Viewers can check the product information on their smartphones based on the information sent from the server, and then click on the link to be taken to the e-commerce site and complete the purchase process.

[1362] The above is the "Mode for Carrying Out the Invention" of the present invention. This system allows viewers to quickly and accurately obtain information about items that interest them, and promotes purchasing behavior.

[1363] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1364] Step 1:

[1365] Viewers tap on items that interest them while watching video content on their device.

[1366] Input: Viewer taps

[1367] Output: Tap location and frame data of the video being watched

[1368] How it works: The device captures and records the location of the tap and the video frame data at that moment. This information is used to make requests to the server.

[1369] Step 2:

[1370] The tapped position information and frame data are sent to the server.

[1371] Input: Tap location and frame data of the video being watched

[1372] Output: HTTP request to the server

[1373] Operation: The device generates and sends an HTTP request to the server to send the collected location information and frame data.

[1374] Step 3:

[1375] The server analyzes the objects in the frame based on the information received and identifies the relevant item.

[1376] Input: Tap location and frame data

[1377] Output: Feature information of the identified item (e.g. color, shape, brand)

[1378] How it works: The server uses an object recognition AI model to analyze the frame data and identify the item of interest based on the transmitted tap location. From this analysis, it extracts feature information about the item.

[1379] Step 4:

[1380] A database is searched based on the characteristic information of the identified item to obtain related product information.

[1381] Input: Feature information of the identified item

[1382] Output: Related product information (product name, price, purchase link, thumbnail image, etc.)

[1383] Operation: The server uses the acquired feature information to search the database to identify and retrieve relevant product information. The search results include the most relevant product information.

[1384] Step 5:

[1385] Based on the generated product information, a prompt sentence is generated using a generative AI model and presented to the viewer.

[1386] Input: Related product information

[1387] Output: Prompt (e.g., "While watching the video, I noticed the blue dress the actress was wearing. If you tap on the item, more information will appear on the screen.")

[1388] How it works: The server uses a generative AI model to convert the acquired product information into detailed prompts that are used to communicate product details to the viewer.

[1389] Step 6:

[1390] The server sends product information and a prompt to the device.

[1391] Input: Prompt statement and product information

[1392] Output: Sending data to the viewer's device

[1393] How it works: The server generates a response to send the generated prompt and product information to the viewer's device, and sends it to the device using the HTTP protocol.

[1394] Step 7:

[1395] The terminal displays the received product information and prompt text on the screen.

[1396] Input: Received product information and prompt text

[1397] Output: Product information and prompt text displayed on the screen

[1398] Operation: The terminal analyzes the product information and prompt text received from the server and displays it on the screen in a format that is easy for the viewer to see.

[1399] Step 8:

[1400] Viewers click on the provided link to view more information.

[1401] Input: Product information link

[1402] Output: Transition to EC site and display of detailed information

[1403] How it works: When a viewer clicks on a product link displayed on the screen, the device's browser will redirect to the corresponding e-commerce site, where they can view detailed information and complete the purchase process.

[1404] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1405] The present invention is a system that easily obtains information about items that interest a viewer while they are watching video content on their device, and by combining this with an emotion engine, provides information and promotes purchasing behavior that is tailored to the viewer's emotions. This system starts when the viewer taps on an item of interest on the device screen, and includes a process in which a server recognizes the item and provides related product information. The system also uses the emotion engine to analyze the viewer's emotions and optimize the information provided.

[1406] Server-side processing

[1407] 1. Running the video analysis module:

[1408] The server receives the video content and begins analyzing it frame by frame, using an object recognition AI model to identify items within the frame.

[1409] 2. Database search of identification results:

[1410] Based on the information of the recognized object (for example, color, shape, brand logo, etc. for clothing), the e-commerce site's database is searched to identify related product information.

[1411] 3. Ranking of specific products:

[1412] Based on the search results, the AI ​​ranks the most suitable products based on the viewer's past browsing and purchasing history.

[1413] 4. Running the Emotion Engine:

[1414] The server analyzes the viewer's facial expressions and voice data using an emotion engine to identify the viewer's emotional state.

[1415] 5. Emotion-based filtering:

[1416] Filter or rank product information based on the viewer's emotional state.

[1417] 6. Provision of Information:

[1418] Information about the tapped item is instantly sent to the device, and data is prepared that links to the details page of the product the viewer needs.

[1419] Terminal side processing

[1420] 1. Listening for tap events:

[1421] The device monitors the viewer's tap events and sends the location information to the server when a specific item is tapped.

[1422] 2. Sending a request to the server:

[1423] A request is sent to the server along with the frame data of the tap position, and this information is used by the server's analysis module.

[1424] 3. Information Receipt and Display:

[1425] It receives product information sent from the server and displays related information on the screen while the user is viewing the content. It improves user convenience by displaying links to related e-commerce sites and thumbnail images of the products.

[1426] User processing

[1427] 1. Item tap:

[1428] The viewer taps on an item that interests them (e.g., an actress's clothing). This action records the location where the tap was made on the device.

[1429] 2. Information verification:

[1430] Check the product information displayed on your device. Click on the link to the related e-commerce site to view more information.

[1431] 3. Purchasing decision:

[1432] They decide on their purchasing behavior based on the information provided, and if they like the product, they complete the purchase process through the e-commerce site.

[1433] Specific examples

[1434] 1. User Scenario:

[1435] While watching a drama starring an actress on his smartphone, a user becomes interested in the dress the actress is wearing and taps on the dress.

[1436] 2. Tap event occurs:

[1437] The device detects the tap event and sends the tap position and frame information to the server.

[1438] 3. Parsing on the server:

[1439] The server analyzes the items in the frame and recognizes the dress. Based on the recognition results, it searches the e-commerce site database to identify the most suitable dress.

[1440] 4. Running the Emotion Engine:

[1441] The server analyzes the user's emotions from facial expressions and voice data and determines, for example, whether the user is excited.

[1442] 5. Emotional information:

[1443] Based on the information from the emotion engine, the server provides preferentially information on products that suit the user's state of excitement.

[1444] 6. Information provision:

[1445] The server sends the details of the dress and a purchase link to the device, which then displays the details of the dress to the user based on the received information.

[1446] 7. Purchasing behavior:

[1447] The user checks the displayed information, and if they like the dress, they click the provided link to go to the e-commerce site and complete the purchase process.

[1448] According to the above-described embodiment, the present invention can quickly and easily obtain information about items that interest a viewer, and can further promote purchasing behavior by displaying information according to the viewer's emotions.

[1449] The processing flow will be explained below.

[1450] Step 1:

[1451] A user watches video content on a device and taps an item that interests them (e.g., an actress's clothing). This action records the location of the tap on the device.

[1452] Step 2:

[1453] The device detects a touch event and obtains the touch location, the current video frame data, and a timestamp, including the time at which the operation occurred.

[1454] Step 3:

[1455] The device sends the tap location, frame data, and timestamp to the server in a request that also includes the necessary metadata.

[1456] Step 4:

[1457] Based on the tap location information and frame data received by the server, a video analysis module is used to analyze the objects in the frame, using an object recognition AI model to identify the item closest to the tapped location.

[1458] Step 5:

[1459] The server extracts the characteristic information (e.g., color, shape, brand, pattern) of the identified item and searches the e-commerce site's database to obtain related product information.

[1460] Step 6:

[1461] The server retrieves relevant product information from the search results and ranks the most suitable products based on the viewer's past browsing and purchasing history, taking user preferences into account in the process.

[1462] Step 7:

[1463] The server runs an emotion engine that analyzes the viewer's facial and voice data to determine their emotions, in this case emotional states such as excitement, joy, and surprise.

[1464] Step 8:

[1465] The server further filters the ranked product information based on the user's emotional state, prioritizing products that best suit the user's current emotions.

[1466] Step 9:

[1467] The server sends the identified product information (product name, price, sales site link, image, etc.) to the device. This information is optimized based on emotions.

[1468] Step 10:

[1469] The product information received by the device is displayed on the screen in a format that is intuitive and easy for viewers to understand, including pop-ups and overlays.

[1470] Step 11:

[1471] Users review the displayed product information and click on the link to the product detail page of the product they are interested in. This action makes it easier to obtain additional information.

[1472] Step 12:

[1473] The device will redirect to an e-commerce site where the user can view detailed information, allowing them to make a final purchasing decision.

[1474] Example 2

[1475] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1476] Conventional video viewing systems lack a means for viewers to quickly and easily obtain information about items that interest them within video content. Furthermore, there is no method for providing information or encouraging purchasing behavior based on the viewer's emotional state. This makes it difficult to effectively stimulate viewers' purchasing desire and provide an optimal purchasing experience.

[1477] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for tapping an item in which a viewer is interested, a means for transmitting tapped position information and frame data to the server, a means for analyzing an object in the frame based on the transmitted information and identifying the relevant item, a means for searching a database based on characteristic information of the identified item and acquiring related product information, a means for analyzing facial expressions and voice data of the viewer and identifying the emotional state, and a means for filtering or ranking product information based on the identified emotional state. This makes it possible to quickly and easily acquire information about items in which the viewer is interested, and to optimally display information and promote purchases according to the viewer's emotions.

[1478] A "viewer" is a user who is viewing video content on a device.

[1479] A "device" is an electronic device (e.g., smartphone, tablet, computer) that displays video content and can be operated by a user.

[1480] "Video content" means digital media containing a combination of moving images and sounds that convey information through visual and auditory means.

[1481] An "item" refers to an object or person that appears in video content and is an object of interest to viewers.

[1482] "Tap" means that the viewer selects an item by touching the device's touchscreen.

[1483] "Tap location information" is the coordinate data when a viewer taps an item on the device screen.

[1484] "Frame data" refers to information about each frame (still image) of video content, and is data that represents video at a specific point in time.

[1485] A "server" is a central processing unit that receives data from devices and performs analytical processing and database searches.

[1486] "Object analysis means" refers to the technology or algorithms used by the server to identify objects or items present in each frame of video content (e.g., object recognition AI models).

[1487] The "means for identifying the relevant item" is a method for finding the item that the viewer tapped on from among the analyzed objects.

[1488] The "means for searching a database" is a method for querying a database in which various product information is stored based on the characteristic information of the identified item.

[1489] "Related Product Information" is detailed information (e.g., price, description, images, links) about products that are similar or match the identified item.

[1490] The "means for displaying to the viewer" refers to a method for visually presenting the acquired product information on a device.

[1491] "Means for analyzing facial and / or audio data" means the techniques or algorithms used to infer the emotional state of a viewer from their facial expressions or audio (e.g., emotion engine).

[1492] "Emotional state" is data that reflects the type and intensity of a viewer's emotion, including joy, excitement, sadness, etc.

[1493] A "filtering or ranking means" is a method for sorting or prioritizing retrieved product information based on the identified emotional state.

[1494] The present invention is a system that allows a viewer to quickly and easily obtain information about items that interest them while watching video content on a device, and provides information that corresponds to the viewer's emotions and promotes purchasing behavior.

[1495] The overall operation of the system involves three entities: the server, the device, and the viewer. We will explain how each entity interacts and processes information.

[1496] Server-side processing

[1497] First, we will explain the server-side processing. Specifically, it has the following functions:

[1498] 1. Video Analysis Module:

[1499] The server splits the video content received from the user into frames and identifies items within the frames using an object recognition AI model (e.g., YOLO or Detectron2). Through this analysis, the server identifies items that may be of interest to the viewer.

[1500] 2. Database Search:

[1501] Based on the analyzed item's attribute information (color, shape, brand logo, etc.), the server queries the e-commerce site's database (e.g., Amazon Product Advertising API, Rakuten product database) to obtain related product information.

[1502] 3. Ranking Algorithm:

[1503] Based on the acquired product information, the server takes into account the viewer's past browsing and purchasing history and uses AI algorithms (such as collaborative filtering or content-based filtering) to rank the products that are best suited to the viewer.

[1504] 4. Sentiment Analysis Module:

[1505] The server acquires the viewer's facial expression and voice data and analyzes it using an emotion engine (e.g., Microsoft Azure Emotion API, IBM Watson Tone Analyzer) to determine their real-time emotional state (excitement, joy, calmness, etc.).

[1506] 5. Emotion-based filtering:

[1507] The server filters or re-ranks product information based on the viewer's emotional state, for example, prioritizing special offers and limited edition products for an excited viewer.

[1508] 6. Provision of Information:

[1509] The server immediately transmits the filtered product information to the terminal and provides a link to the detailed page of the product the viewer needs.

[1510] Terminal side processing

[1511] Next, the processing on the terminal side will be described.

[1512] 1. Listening for tap events:

[1513] The device monitors the viewer's tapping on the screen while watching a video, acquires the location information of the tap, and sends this location information to the server.

[1514] 2. Data transmission:

[1515] The tapped position information and frame data are sent to the server, which uses this information in the server's video analysis module.

[1516] 3. Information Receipt and Display:

[1517] It receives product information sent from the server and displays related information on the screen while the user is watching. Specifically, it displays product information pop-ups, links to related e-commerce sites, thumbnail images of the products, and more.

[1518] User processing

[1519] Finally, the processing on the user side will be described.

[1520] 1. Item tap:

[1521] The user taps on an item of interest (e.g., the actress's clothing) in the video they are watching. This action records the location of the tap on the device.

[1522] 2. Information verification:

[1523] The user checks the product information displayed on the device, clicks on the link to the related e-commerce site, and views the product details page.

[1524] 3. Purchasing decision:

[1525] The user decides on a purchasing decision based on the information provided, and if they like the product, they complete the purchase process through the e-commerce site.

[1526] Specific examples

[1527] Next, a specific example of the operation of the present invention will be described.

[1528] 1. User Scenario:

[1529] While watching a drama starring an actress on his smartphone, a user becomes interested in the dress the actress is wearing and taps on the dress.

[1530] 2. Tap event occurs:

[1531] The device detects the tap event and sends the tap position and frame information to the server. Specifically, it sends the tap coordinates and the frame data at that time.

[1532] 3. Parsing on the server:

[1533] The server analyzes the items in the received frame and recognizes the dress. Based on the recognition results, it searches the e-commerce site database to identify the most suitable dress.

[1534] 4. Running the Emotion Engine:

[1535] The server analyzes the user's emotions from facial expressions and voice data, and determines, for example, whether the user is in an excited state.

[1536] 5. Emotional information:

[1537] The server provides information on products that suit the user's state of excitement with priority based on information from the emotion engine.

[1538] 6. Information provision:

[1539] The server sends detailed information about the dress and a link to purchase it to the device, which then displays the details of the dress to the user based on the information received.

[1540] 7. Purchasing behavior:

[1541] The user checks the displayed information, and if they like the dress, they click on the provided link to go to the e-commerce site and complete the purchase process.

[1542] Example prompt:

[1543] "If I want to know about the dress an actress is wearing, how can the server provide this information and explain the process for the user to purchase it?"

[1544] As described above, the system of the present invention is designed to quickly and easily obtain information about items that interest viewers, and to display information optimally in response to the viewers' emotions and promote purchases.

[1545] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1546] Step 1:

[1547] Listening for touch events

[1548] The device monitors the viewer's tapping on the screen while watching a video. The input is the viewer's tapping on the touchscreen. Specifically, the device obtains the tapped position information (coordinate data) and the frame data indicating which frame was being displayed. The output is the tapped position information and frame data.

[1549] Step 2:

[1550] Sending a request to the server

[1551] The device sends the acquired tap position information and frame data to the server. The inputs include tap position information and frame data. Specifically, the device sends this information to the server in the form of an HTTP request or similar. The output includes the tap position information and frame data sent to the server.

[1552] Step 3:

[1553] Running the video analysis module

[1554] The server uses the received frame data to run an object recognition AI model (e.g., YOLO or Detectron2) and analyze the objects in the frame. The input is the frame data. Specifically, the server analyzes the frame and identifies the item. The output is information about the analyzed item.

[1555] Step 4:

[1556] Database search of identification results

[1557] The server searches the e-commerce site's database based on the analyzed item's characteristic information (color, shape, brand logo, etc.). The analyzed item information is the input. Specifically, the server calls the API to retrieve related product information. The retrieved product information is the output.

[1558] Step 5:

[1559] Ranking of specific products

[1560] The server uses AI algorithms (such as collaborative filtering or content-based filtering) to rank the acquired product information. The inputs are the acquired product information and the viewer's past browsing history and purchase history. Specifically, the server uses AI to rank the products. The output is the ranked product information.

[1561] Step 6:

[1562] Running the Emotion Engine

[1563] The server acquires the viewer's facial expression and voice data and analyzes it using an emotion engine (e.g., Microsoft Azure Emotion API, IBM Watson Tone Analyzer). The input is the viewer's facial expression and voice data. Specifically, the server analyzes the emotion data and identifies the viewer's emotional state. The output is the analyzed emotional state.

[1564] Step 7:

[1565] Sentiment-based filtering

[1566] The server filters or re-ranks the product information based on the analyzed emotional state. The inputs include the analyzed emotional state and the ranked product information. As a specific operation, the server sorts or prioritizes the product information according to the emotional state. The output includes the filtered or re-ranked product information.

[1567] Step 8:

[1568] Providing information

[1569] The server transmits the filtered or re-ranked product information to the terminal. The input includes the filtered or re-ranked product information. As a specific operation, the server converts the product information into a data packet and transmits it to the terminal. The output includes the product information transmitted to the terminal.

[1570] Step 9:

[1571] Information reception and display

[1572] The device receives product information sent from the server and displays related information on the screen being viewed. The input is the product information sent from the server. In concrete terms, the device presents the product information to the user in the form of a pop-up display or related links. The output is the product information displayed in a format that the viewer can check.

[1573] Step 10:

[1574] Information verification and purchasing behavior

[1575] The user checks the product information displayed on the device and, if interested, proceeds with the purchase. The input is the product information displayed on the device. The specific operation is for the user to click on a link to the relevant e-commerce site and proceed with the purchase. The output is a result indicating whether the purchase was completed.

[1576] Through the above steps, viewers can easily obtain information about items that interest them, and information is displayed according to their emotions, promoting purchasing behavior.

[1577] (Application example 2)

[1578] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1579] Currently, there are insufficient ways for viewers to easily obtain information about items they are interested in while watching video content.In addition, information is not provided according to the viewer's emotional state, making it difficult to suggest optimal product information.

[1580] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1581] In this invention, the server includes means for tapping on an object of interest to a viewer while watching video content on a device, means for transmitting tapped position information and frame data to the server, means for analyzing objects in the frame based on the transmitted information and identifying the relevant object, means for searching a database based on characteristic information of the identified object and acquiring related product information, means for transmitting the acquired product information to the device and displaying it to the viewer, means for analyzing the viewer's facial expressions and voice data using an emotion engine and identifying the viewer's emotional state, and means for filtering or ranking the product information to be provided based on the viewer's emotional state. This allows the viewer to quickly and easily obtain information on items that interest them while watching a video, and further makes it possible to provide optimal product information according to the viewer's emotional state.

[1582] A "viewer" is a person who watches video content.

[1583] A "device" is an electronic device that displays video content and allows viewers to interact with it.

[1584] "Video content" refers to all media content provided in video format.

[1585] An "object" is an object that a viewer becomes interested in while watching video content and tries to obtain information by tapping on it.

[1586] A "tap" is the act of touching a specific location on a device screen once with a finger or stylus.

[1587] "Frame data" refers to each still image data that constitutes video content.

[1588] A "server" is a computer system that receives a request over a network, executes the specified process, and returns a response.

[1589] "Analyzing an object" means recognizing a specific object from frame data in video content and extracting information about it.

[1590] "Feature information" refers to attribute information for identifying an object, such as the object's color, shape, brand logo, etc.

[1591] A "database" is a collection of digital information organized to allow efficient storage and retrieval of specific information.

[1592] "Product Information" refers to detailed information about products and services related to the subject matter.

[1593] An "emotion engine" is a software module that analyzes the viewer's emotional state and optimizes the information provided based on the results.

[1594] "Emotional state" refers to data that represents the viewer's current psychological and physiological state.

[1595] "Filtering" refers to the act of sorting and prioritizing the information provided.

[1596] "Ranking" is the act of assigning a rank to information or objects based on certain criteria.

[1597] The present invention is a system that easily obtains information about objects of interest to viewers while they are watching video content on their devices, and provides optimal product information according to the viewer's emotional state. The system for implementing the present invention consists of a server, a terminal, and user interactions.

[1598] Server Processing

[1599] The server consists of the following modules:

[1600] Object Recognition Module: Using TensorFlow and OpenCV, this module analyzes and identifies objects within frames of video content. This module recognizes objects based on the transmitted location information and frame data, and extracts their feature information.

[1601] Database search module: Using MySQL or MongoDB, this module searches the database for relevant product information based on the characteristics of the identified object. This module quickly retrieves and ranks product information.

[1602] Sentiment analysis engine: Using Google Cloud Vision and Amazon Rekognition, the engine analyzes the viewer's emotional state from facial expressions and voice data, and filters or ranks the product information provided based on the analysis results.

[1603] Terminal handling

[1604] The terminal has the following features:

[1605] Monitoring tap events: Events that occur when a viewer taps the screen of the device are monitored, and the location information and frame data are sent to the server.

[1606] Communication with the server: When a tap event is detected, the information is sent to the server as a request and product information sent from the server is received.

[1607] Information display: Product information obtained from the server is displayed on the screen while the content is being viewed, and related product links and thumbnail images are provided.

[1608] User Action

[1609] The user interacts with:

[1610] Item Tap: Tap on an object that interests you while watching and send its location information.

[1611] Check information: Check the product information displayed on your device and click on the link for more information or to purchase.

[1612] Purchasing behavior: Select a product based on the information provided and complete the purchase process on the e-commerce site.

[1613] Specific examples

[1614] For example, suppose a user is watching a movie on a media distribution platform and is interested in the clothing worn by the lead actor. In this case, the user taps on the device screen. This tap event is detected by the device, and the relevant location information and frame data are sent to the server. The server identifies the clothing based on this data and searches for related product information from a database based on its feature information. Furthermore, an emotion analysis engine analyzes the user's emotional state (e.g., excitement or curiosity), ranks the product information accordingly, and sends it to the device. The device displays the results to the user, who can click on the provided link to check more information and make a purchase.

[1615] Prompt Sentence Examples

[1616] While a user is watching a movie on their smartphone, they tap on an item they are interested in. Can you give us an example of a system that retrieves information about that item and provides relevant product information, taking into account the user's emotional state?

[1617] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1618] Step 1:

[1619] User interaction and tap events

[1620] While a user is watching video content on a device, they tap on an object that interests them.

[1621] Input: User tap position information and video frame data

[1622] What happens: The device's touchscreen detects the user's taps and captures their location.

[1623] Step 2:

[1624] Sending tap position

[1625] The terminal transmits the tap position information and the associated frame data to the server.

[1626] Input: User tap position information and video frame data

[1627] Output: Tap position information and frame data are sent to the server.

[1628] Specific operation: The device detects a touch event and sends the coordinates and current frame data to the server as an HTTP request.

[1629] Step 3:

[1630] Analysis of objects in the frame

[1631] The server analyzes the objects in the frame based on the transmitted location information and frame data, and identifies the relevant target.

[1632] Input: Tap position information and video frame data

[1633] Output: Feature information of identified objects

[1634] Specific operation: The object recognition module (TensorFlow or OpenCV) on the server recognizes the object around the tap position in the frame and extracts its feature information.

[1635] Step 4:

[1636] Database search

[1637] The server searches the database based on the characteristic information of the identified object and obtains related product information.

[1638] Input: Feature information of the identified object

[1639] Output: List of product details

[1640] Specific operation: Using the feature information, query the database (MySQL or MongoDB) to retrieve the corresponding product information.

[1641] Step 5:

[1642] Emotional state analysis

[1643] The server uses an emotion analysis engine to identify the viewer's emotional state based on the user's facial expression data and voice data.

[1644] Input: User's facial expression data and voice data

[1645] Output: Emotional state assessment result

[1646] Specific operation: Analyzes the user's emotional state using APIs such as Google Cloud Vision and Amazon Rekognition.

[1647] Step 6:

[1648] Filtering and ranking product information

[1649] The server filters or ranks the product information based on the analyzed emotional state.

[1650] Input: List of product information, emotional state evaluation results

[1651] Output: Filtered or ranked product information

[1652] Specific behavior: Select and prioritize the most relevant product information based on emotional state.

[1653] Step 7:

[1654] Sending and Displaying Information

[1655] The server transmits the filtered product information to the terminal, which displays it on the screen being viewed.

[1656] Input: Filtered or ranked product information

[1657] Output: Display information on the user's device (product image, link to details, etc.)

[1658] Specific operation: The server sends the generated product information back to the device as an HTTP response, and the device receives the information and displays it on the screen.

[1659] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1660] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1661] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1662] [Fourth embodiment]

[1663] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1664] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1665] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1666] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1667] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1668] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1669] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1670] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1671] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1672] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1673] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1674] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1675] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1676] The present invention provides a system that allows viewers to easily obtain information about items they are interested in while watching video content on their device, leading to purchasing behavior. This system starts when the viewer taps on an item they are interested in on the device screen, and includes a process in which a server recognizes the item and provides related product information.

[1677] Server-side processing

[1678] 1. Running the video analysis module:

[1679] The server receives the video content and begins analyzing it frame by frame, using an object recognition AI model to identify items within the frame.

[1680] 2. Database search of identification results:

[1681] Based on the information of the recognized object (for example, color, shape, brand logo, etc. for clothing), the e-commerce site's database is searched to identify related product information.

[1682] 3. Ranking of specific products:

[1683] Based on the search results, the AI ​​ranks the most suitable products based on the viewer's past browsing and purchasing history.

[1684] 4. Provision of Information:

[1685] Information about the tapped item is instantly sent to the device, and data is prepared that links to the details page of the product the viewer needs.

[1686] Terminal side processing

[1687] 1. Listening for tap events:

[1688] The device monitors the viewer's tap events and sends the location information to the server when a specific item is tapped.

[1689] 2. Sending a request to the server:

[1690] A request is sent to the server along with the frame data of the tap position, and this information is used by the server's analysis module.

[1691] 3. Information Receipt and Display:

[1692] It receives product information sent from the server and displays related information on the screen while the user is viewing the content. It improves user convenience by displaying links to related e-commerce sites and thumbnail images of the products.

[1693] User processing

[1694] 1. Item tap:

[1695] The viewer taps on an item that interests them (e.g., the clothes worn by an actress). This action records the location where the tap was made on the device.

[1696] 2. Information verification:

[1697] Check the product information displayed on your device. Click on the link to the related e-commerce site to view more information.

[1698] 3. Purchasing decision:

[1699] They decide on their purchasing behavior based on the information provided, and if they like the product, they complete the purchase process through the e-commerce site.

[1700] Specific examples

[1701] 1. User Scenario:

[1702] While watching a drama starring an actress on his smartphone, a user becomes interested in the dress the actress is wearing and taps on the dress.

[1703] 2. Tap event occurs:

[1704] The device detects the tap event and sends the tap position and frame information to the server.

[1705] 3. Parsing on the server:

[1706] The server analyzes the items in the frame and recognizes the dress. Based on the recognition results, it searches the e-commerce site database to identify the most suitable dress.

[1707] 4. Information provision:

[1708] The server sends the details of the dress and a purchase link to the device, which then displays the details of the dress to the user based on the received information.

[1709] 5. Purchasing behavior:

[1710] The user checks the displayed information, and if they like the dress, they click the provided link to go to the e-commerce site and complete the purchase process.

[1711] In the above-described form, the present invention allows viewers to quickly and easily obtain information about items that interest them, thereby promoting purchasing behavior.

[1712] The processing flow will be explained below.

[1713] Step 1:

[1714] A user watches video content on a device and taps on an item of interest (e.g., an actress's clothing).

[1715] Step 2:

[1716] The device detects the touch event and retrieves the touch location and the current video frame data, including a specific timestamp in the video.

[1717] Step 3:

[1718] The device sends the tap location, frame data, and timestamp to the server, along with any necessary metadata.

[1719] Step 4:

[1720] Based on the tap position information and frame data received by the server, a video analysis module is used to analyze the object in the corresponding frame.

[1721] Step 5:

[1722] The server uses the analysis to identify the item closest to the tapped location, which is performed by an object recognition AI model.

[1723] Step 6:

[1724] The server extracts the characteristic information of the identified item (e.g., color, shape, brand, pattern) and searches the e-commerce site's database.

[1725] Step 7:

[1726] The server retrieves relevant product information from the database and ranks the most suitable products by referring to the viewer's past browsing and purchasing history.

[1727] Step 8:

[1728] The server sends the ranked product information (product name, price, sales site link, image, etc.) to the terminal.

[1729] Step 9:

[1730] The product information received by the device is displayed on the screen in a pop-up or overlay format.

[1731] Step 10:

[1732] The user checks the displayed product information and clicks on the link to the product detail page of the product that interests them.

[1733] Step 11:

[1734] The device will redirect to the EC site, where the user can view detailed information and then complete the purchase procedure.

[1735] Example 1

[1736] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1737] When viewers watch video content on their devices, there is a demand for a system that can quickly and easily obtain information about items that interest them and directly link this information to purchasing behavior. Conventional systems have the problem that the process from finding a product that interests a viewer to purchasing it is complicated and time-consuming. Furthermore, they lack a mechanism for suggesting optimal products based on the viewer's past preferences and browsing history.

[1738] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1739] In this invention, the server includes means for analyzing objects in the frame based on the transmitted information and identifying the relevant item, means for searching a database based on attribute information of the identified item and acquiring related product information, and means for transmitting the acquired product information to the device and displaying it to the viewer. This enables a viewer to quickly acquire product information related to an item that interests them in the video, and receive appropriate product suggestions, simply by tapping on that item.

[1740] A "viewer" is someone who watches video content on a device and interacts with items of interest.

[1741] A "device" is a device that a viewer uses to watch video content, and includes, for example, a smartphone, a tablet, a PC, and the like.

[1742] "Video content" means media, including video and audio, that is played by a viewer on a device.

[1743] "Item" refers to an object or subject that appears within video content.

[1744] "Tapping means" refers to a device or method by which a viewer operates the touchscreen of a device to select a particular item.

[1745] "Tap location information" refers to the coordinate information when a viewer taps an item on the device screen.

[1746] "Frame data" refers to image data of a specific moment (frame) in video content.

[1747] "Server" refers to a computer system that receives and processes data sent from a viewer's device.

[1748] "Means for transmitting" refers to a method or apparatus for communicating data between a device and a server.

[1749] "Means for analyzing" refers to a device or method for identifying an object in video content based on the transmitted information and extracting its attribute information.

[1750] "Means for identifying" refers to a device or method that identifies the relevant item from the analyzed data.

[1751] "Attribute information" refers to characteristic information about the identified item (e.g., color, shape, brand logo, etc.).

[1752] The term "means for searching a database" refers to a device or method for obtaining related product information from a database based on the attribute information of a specified item.

[1753] "Product Information" refers to detailed information about related products (e.g., price, image URL, purchase link, etc.).

[1754] "Means for transmitting and displaying to the viewer" refers to a method or apparatus for transmitting product information obtained from the server to the viewer's device and displaying it on the viewer's screen.

[1755] "Past browsing history" and "purchase history" refer to historical information about content that a viewer has viewed or products that they have purchased in the past.

[1756] The term "ranking means" refers to a device or method that evaluates and prioritizes relevant product information based on the viewer's past browsing history and purchasing history.

[1757] The "means for accepting interaction" refers to a method or device that allows a viewer to perform an operation to view detailed information from the provided product information.

[1758] "Means for transitioning to an e-commerce site" refers to a device or method that provides a link to an e-commerce site and directs the viewer to that site for more information.

[1759] The present invention provides a system that allows viewers to easily obtain information about items they are interested in while watching video content on their device, leading to purchasing behavior. This system starts when the viewer taps on an item they are interested in on the screen of their device, and includes a process in which a server recognizes the item and provides related product information.

[1760] Server-side processing

[1761] The server receives video content sent from the viewer's device and analyzes the video frame by frame using an object recognition AI model (e.g., YOLOv5 or Mask R-CNN). The server identifies the items present in each frame and extracts their attributes (e.g., color, shape, brand logo).

[1762] Based on the attributes of the identified objects, an SQL query is executed to search the e-commerce site's database to retrieve matching related products. The server also uses machine learning models (e.g., collaborative filtering and deep learning) to rank the best products based on the viewer's past browsing and purchasing history.

[1763] The server constructs the identified product information and ranking results in JSON format and sends them to the terminal using a RESTful API.

[1764] Terminal side processing

[1765] The device uses a touchscreen event listener to monitor events when a specific item in the video is tapped. The touch location information is captured and temporarily saved.

[1766] This location information and the current frame data are sent to the server using an HTTP POST request. The product information received from the server is analyzed and relevant information is displayed on the screen while the user is watching. Product information includes thumbnail images, prices, purchase links, etc.

[1767] User processing

[1768] Viewers tap on an item of interest (for example, clothing worn by an actress in a drama). The location of the tap is recorded on the device and sent to the server. The viewer checks the product information sent from the server and clicks on a link to the related e-commerce site to view more information. If they find a product they like, they can complete the purchase process through the e-commerce site.

[1769] Specific examples

[1770] For example, a user might be watching a TV drama on their smartphone and become interested in the dress an actress is wearing. When the user taps on the dress, the device detects the tap event and sends the tap location and frame information to the server. The server then analyzes the items in the frame and recognizes the target dress. Based on the recognition results, the database of the e-commerce site is searched to identify the most suitable dress.

[1771] The server sends detailed information about the dress and a link to purchase it to the device, which then displays it to the viewer. The user checks the information displayed, and if they like the dress, they click the provided link to go to the e-commerce site and complete the purchase process.

[1772] This system allows viewers to quickly and easily obtain information about items that interest them, thereby promoting purchasing behavior.

[1773] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1774] Step 1:

[1775] Monitor tap events (device)

[1776] The device monitors events when the viewer taps on a specific item in the video, and uses a touchscreen event listener to obtain the location information when the viewer taps on the item.

[1777] Input: Viewer taps.

[1778] Data processing: Obtain the coordinates of the tapped location and temporarily save them.

[1779] Output: Coordinate information of the tap position.

[1780] Specific behavior:

[1781] Set a touch event listener on the video player.

[1782] Get the tapped coordinate position and save it in a variable.

[1783] Step 2:

[1784] Send tap information to server (device)

[1785] The device sends the acquired tap position information and the current frame data to the server using an HTTP POST request.

[1786] Input: Tap location and frame data.

[1787] Data processing: Convert tap position information and frame data into JSON format.

[1788] Output: The HTTP POST request sent to the server.

[1789] Specific behavior:

[1790] Build tap position information and frame data in JSON format.

[1791] Creates an HTTP POST request and sends it to the server.

[1792] Step 3:

[1793] Video analysis (server)

[1794] The server analyzes the objects in the frame based on the received tap information and frame data, and uses an object recognition AI model (e.g., YOLOv5 or Mask R-CNN) to identify the items and extract their attribute information.

[1795] Input: Tap position information and frame data.

[1796] Data processing: Input the frame data into an object recognition AI model to extract specific item attributes (color, shape, brand logo, etc.).

[1797] Output: Attribute information of the recognized item.

[1798] Specific behavior:

[1799] The frame data stored on the server is input into the object recognition AI model.

[1800] Temporarily save the attribute information of the item obtained from the model.

[1801] Step 4:

[1802] Database search (server)

[1803] The server searches the e-commerce site's database based on the attribute information of the identified item to retrieve related product information. It executes an SQL query to retrieve matching product information from the database.

[1804] Input: Item attribute information.

[1805] Data processing: Generate SQL queries and retrieve product information from the database.

[1806] Output: Retrieved product information.

[1807] Specific behavior:

[1808] Generates an SQL query using the item's attribute information.

[1809] Query the database to retrieve related product information.

[1810] The acquired product information is temporarily saved.

[1811] Step 5:

[1812] Product information ranking (server)

[1813] The server ranks the product information obtained based on the viewer's past browsing and purchasing history, and uses machine learning models (e.g., collaborative filtering and deep learning) to select the best products.

[1814] Input: Retrieved product information, viewer's past browsing and purchasing history.

[1815] Data processing: Input data into a machine learning model and rank it.

[1816] Output: Ranked product information.

[1817] Specific behavior:

[1818] Obtain past viewer data from the database.

[1819] Product information and historical data are input into a machine learning model to obtain ranking results.

[1820] Temporarily save the ranking results.

[1821] Step 6:

[1822] Product information sent to server (server)

[1823] The server sends the ranked product information to the terminal and returns the product information in JSON format using a RESTful API.

[1824] Input: Ranked product information.

[1825] Data processing: Build product information in JSON format.

[1826] Output: The HTTP response sent to the device.

[1827] Specific behavior:

[1828] Build product information in JSON format.

[1829] Use the RESTful API to generate an HTTP response and send it to the device.

[1830] Step 7:

[1831] Display product information (device)

[1832] The terminal analyzes the product information received from the server and displays related information on the screen being viewed.

[1833] Input: The HTTP response from the server.

[1834] Data processing: Analyze JSON data and extract product information.

[1835] Output: Product information displayed on the screen.

[1836] Specific behavior:

[1837] Parse the JSON data received from the server.

[1838] Display the extracted product information on the screen (e.g., by placing a thumbnail image or purchase link).

[1839] Step 8:

[1840] Check detailed information (user)

[1841] The user checks the product information displayed on the terminal and clicks on the link to view more detailed information.

[1842] Input: View product information.

[1843] Data processing: Clicking on a link.

[1844] Output: Show detailed information.

[1845] Specific behavior:

[1846] Check the detailed information of the displayed product.

[1847] Click on the link to access the e-commerce site, if desired.

[1848] Step 9:

[1849] Purchasing decision (user)

[1850] The user decides on a purchase based on the information provided, and if they find a product they like, they complete the purchase process through the e-commerce site.

[1851] Input: Display product details.

[1852] Data processing: Operations to carry out purchase procedures.

[1853] Output: Product purchase completed.

[1854] Specific behavior:

[1855] Check the product details and decide to purchase.

[1856] Proceed through a purchase on an e-commerce site.

[1857] (Application example 1)

[1858] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1859] In today's content distribution services, it is difficult for viewers to quickly and accurately obtain information about items they are interested in while watching a video, leading to immediate purchase behavior. There is also a need for a system that allows viewers to obtain detailed information about items along with related products. Given this situation, a means is needed to efficiently satisfy viewers' interests and encourage purchasing behavior without compromising the viewing experience.

[1860] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1861] In this invention, the server includes means for tapping on an item of interest to a viewer while watching video content on a device, means for transmitting tapped position information and frame data to the server, means for analyzing each video frame using an object recognition AI model to identify the relevant item, means for searching a database based on feature information of the identified item to obtain related product information, means for transmitting the obtained product information to the device and displaying it to the viewer, and means for generating detailed data about the tapped item as a prompt sentence using a generative AI model and presenting it to the viewer. This makes it possible to quickly obtain information about the item being viewed and promote purchasing behavior by providing necessary detailed information while maintaining the viewing experience.

[1862] A "viewer" is a user who watches video content.

[1863] A "device" is a hardware device for viewing video content, and specifically includes smartphones, tablets, and computers.

[1864] "Video content" refers to content that combines video and audio that is viewed by a viewer through a device.

[1865] An "item" is an object or element that appears in video content and is an object that may interest viewers.

[1866] A "tap" is an action in which the viewer touches the device's touch panel and identifies the location.

[1867] "Location information" refers to the coordinate data on the screen of the device that was tapped.

[1868] "Frame data" is still image data that represents a specific moment in video content.

[1869] A "server" is a computer system that receives, processes, and analyzes data sent from devices.

[1870] An "object recognition AI model" is an artificial intelligence algorithm for identifying specific items within a frame of video content.

[1871] "Feature information" is data relating to the attributes and identifiers of the identified item.

[1872] A "database" is a collection of information in which related product information is stored.

[1873] "Obtained product information" is detailed data about products related to the identified item.

[1874] A "generative AI model" is an artificial intelligence algorithm that generates detailed data about a specific item.

[1875] A "prompt" is a detailed explanatory text generated by a generative AI model to be presented to the viewer.

[1876] "Viewing history" is a record of video content that a viewer has viewed in the past.

[1877] "Purchase history" is a record of products that a viewer has purchased in the past.

[1878] "Ranking" is an evaluation process that ranks related products based on viewing history and purchasing history.

[1879] "Interaction" refers to the operations or inputs that viewers make through a device.

[1880] An "EC site" is a website for conducting electronic commerce.

[1881] The present invention provides a system for promptly and accurately providing information about items of interest to viewers while they are watching video content on their devices, thereby promoting purchasing behavior. An embodiment of this system will be described below.

[1882] System configuration

[1883] This system mainly consists of a server, viewer terminals, and a database.

[1884] 1. Server

[1885] Hardware: General cloud server (e.g. AWS EC2)

[1886] Software: Flask (web framework), OpenCV (image processing library), TensorFlow or PyTorch (object recognition AI model), generative AI model

[1887] Function: The server receives tap events from viewers, analyzes video frames to identify items, retrieves relevant product information from a database, and generates detailed prompts using a generative AI model.

[1888] 2. Terminal

[1889] Hardware: Smartphones, tablets

[1890] Software: Tap event monitoring application, HTTP request library

[1891] Function: The device detects tap events while the viewer is watching video content, sends the location information and frame data to the server, and displays the retrieved product information.

[1892] 3. Database

[1893] Components: A database containing product information for an e-commerce site

[1894] Function: Provides related product information based on the characteristics of the identified item.

[1895] Operation flow

[1896] 1. Viewer Controls

[1897] When a viewer is watching video content on a device, they tap on an item they are interested in. At this time, the device transmits information about the tap position and the corresponding frame data to the server.

[1898] 2. Processing on the server

[1899] Based on the received data, the server uses an object recognition AI model to analyze the items in the frame and identify the relevant item. It then searches a database based on the identified item's characteristics (e.g., color, shape, brand) to retrieve related product information. It then uses a generative AI model to generate a detailed prompt for the relevant product and present it to the viewer. This prompt includes the product name, price, and a purchase link.

[1900] 3. Display on device

[1901] The product information received from the server is displayed on the device, and the viewer can check the provided information. If the viewer is interested, they can be redirected to an e-commerce site for more detailed information or to complete the purchase process.

[1902] Specific examples

[1903] While watching a drama on a smartphone, a viewer becomes interested in the blue dress worn by the actress and taps on that part. At the same time, the device sends its location information and frame data to a server. The server uses an object recognition AI model to analyze the items in the frame and identify the "blue dress." It then searches a database to retrieve related product information (product name: "Blue Dress," price: "10,000 yen," purchase link, thumbnail image). It then uses a generative AI model to generate the following prompt:

[1904] "While watching the video, I noticed a blue dress the actress was wearing. When I tapped on the item, more information was displayed on the screen."

[1905] Viewers can check the product information on their smartphones based on the information sent from the server, and then click on the link to be taken to the e-commerce site and complete the purchase process.

[1906] The above is the "Mode for Carrying Out the Invention" of the present invention. This system allows viewers to quickly and accurately obtain information about items that interest them, and promotes purchasing behavior.

[1907] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1908] Step 1:

[1909] Viewers tap on items that interest them while watching video content on their device.

[1910] Input: Viewer taps

[1911] Output: Tap location and frame data of the video being watched

[1912] How it works: The device captures and records the location of the tap and the video frame data at that moment. This information is used to make requests to the server.

[1913] Step 2:

[1914] The tapped position information and frame data are sent to the server.

[1915] Input: Tap location and frame data of the video being watched

[1916] Output: HTTP request to the server

[1917] Operation: The device generates and sends an HTTP request to the server to send the collected location information and frame data.

[1918] Step 3:

[1919] The server analyzes the objects in the frame based on the information received and identifies the relevant item.

[1920] Input: Tap location and frame data

[1921] Output: Feature information of the identified item (e.g. color, shape, brand)

[1922] How it works: The server uses an object recognition AI model to analyze the frame data and identify the item of interest based on the transmitted tap location. From this analysis, it extracts feature information about the item.

[1923] Step 4:

[1924] A database is searched based on the characteristic information of the identified item to obtain related product information.

[1925] Input: Feature information of the identified item

[1926] Output: Related product information (product name, price, purchase link, thumbnail image, etc.)

[1927] Operation: The server uses the acquired feature information to search the database to identify and retrieve relevant product information. The search results include the most relevant product information.

[1928] Step 5:

[1929] Based on the generated product information, a prompt sentence is generated using a generative AI model and presented to the viewer.

[1930] Input: Related product information

[1931] Output: Prompt (e.g., "While watching the video, I noticed the blue dress the actress was wearing. If you tap on the item, more information will appear on the screen.")

[1932] How it works: The server uses a generative AI model to convert the acquired product information into detailed prompts that are used to communicate product details to the viewer.

[1933] Step 6:

[1934] The server sends product information and a prompt to the device.

[1935] Input: Prompt statement and product information

[1936] Output: Sending data to the viewer's device

[1937] How it works: The server generates a response to send the generated prompt and product information to the viewer's device, and sends it to the device using the HTTP protocol.

[1938] Step 7:

[1939] The terminal displays the received product information and prompt text on the screen.

[1940] Input: Received product information and prompt text

[1941] Output: Product information and prompt text displayed on the screen

[1942] Operation: The terminal analyzes the product information and prompt text received from the server and displays it on the screen in a format that is easy for the viewer to see.

[1943] Step 8:

[1944] Viewers click on the provided link to view more information.

[1945] Input: Product information link

[1946] Output: Transition to EC site and display of detailed information

[1947] How it works: When a viewer clicks on a product link displayed on the screen, the device's browser will redirect to the corresponding e-commerce site, where they can view detailed information and complete the purchase process.

[1948] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1949] The present invention is a system that easily obtains information about items that interest a viewer while they are watching video content on their device, and by combining this with an emotion engine, provides information and promotes purchasing behavior that is tailored to the viewer's emotions. This system starts when the viewer taps on an item of interest on the device screen, and includes a process in which a server recognizes the item and provides related product information. The system also uses the emotion engine to analyze the viewer's emotions and optimize the information provided.

[1950] Server-side processing

[1951] 1. Running the video analysis module:

[1952] The server receives the video content and begins analyzing it frame by frame, using an object recognition AI model to identify items within the frame.

[1953] 2. Database search of identification results:

[1954] Based on the information of the recognized object (for example, color, shape, brand logo, etc. for clothing), the e-commerce site's database is searched to identify related product information.

[1955] 3. Ranking of specific products:

[1956] Based on the search results, the AI ​​ranks the most suitable products based on the viewer's past browsing and purchasing history.

[1957] 4. Running the Emotion Engine:

[1958] The server analyzes the viewer's facial expressions and voice data using an emotion engine to identify the viewer's emotional state.

[1959] 5. Emotion-based filtering:

[1960] Filter or rank product information based on the viewer's emotional state.

[1961] 6. Provision of Information:

[1962] Information about the tapped item is instantly sent to the device, and data is prepared that links to the details page of the product the viewer needs.

[1963] Terminal side processing

[1964] 1. Listening for tap events:

[1965] The device monitors the viewer's tap events and sends the location information to the server when a specific item is tapped.

[1966] 2. Sending a request to the server:

[1967] A request is sent to the server along with the frame data of the tap position, and this information is used by the server's analysis module.

[1968] 3. Information Receipt and Display:

[1969] It receives product information sent from the server and displays related information on the screen while the user is viewing the content. It improves user convenience by displaying links to related e-commerce sites and thumbnail images of the products.

[1970] User processing

[1971] 1. Item tap:

[1972] The viewer taps on an item that interests them (e.g., an actress's clothing). This action records the location where the tap was made on the device.

[1973] 2. Information verification:

[1974] Check the product information displayed on your device. Click on the link to the related e-commerce site to view more information.

[1975] 3. Purchasing decision:

[1976] They decide on their purchasing behavior based on the information provided, and if they like the product, they complete the purchase process through the e-commerce site.

[1977] Specific examples

[1978] 1. User Scenario:

[1979] While watching a drama starring an actress on his smartphone, a user becomes interested in the dress the actress is wearing and taps on the dress.

[1980] 2. Tap event occurs:

[1981] The device detects the tap event and sends the tap position and frame information to the server.

[1982] 3. Parsing on the server:

[1983] The server analyzes the items in the frame and recognizes the dress. Based on the recognition results, it searches the e-commerce site database to identify the most suitable dress.

[1984] 4. Running the Emotion Engine:

[1985] The server analyzes the user's emotions from facial expressions and voice data and determines, for example, whether the user is excited.

[1986] 5. Emotional information:

[1987] Based on the information from the emotion engine, the server provides preferentially information on products that suit the user's state of excitement.

[1988] 6. Information provision:

[1989] The server sends the details of the dress and a purchase link to the device, which then displays the details of the dress to the user based on the received information.

[1990] 7. Purchasing behavior:

[1991] The user checks the displayed information, and if they like the dress, they click the provided link to go to the e-commerce site and complete the purchase process.

[1992] According to the above-described embodiment, the present invention can quickly and easily obtain information about items that interest a viewer, and can further promote purchasing behavior by displaying information according to the viewer's emotions.

[1993] The processing flow will be explained below.

[1994] Step 1:

[1995] A user watches video content on a device and taps an item that interests them (e.g., an actress's clothing). This action records the location of the tap on the device.

[1996] Step 2:

[1997] The device detects a touch event and obtains the touch location, the current video frame data, and a timestamp, including the time at which the operation occurred.

[1998] Step 3:

[1999] The device sends the tap location, frame data, and timestamp to the server in a request that also includes the necessary metadata.

[2000] Step 4:

[2001] Based on the tap location information and frame data received by the server, a video analysis module is used to analyze the objects in the frame, using an object recognition AI model to identify the item closest to the tapped location.

[2002] Step 5:

[2003] The server extracts the characteristic information (e.g., color, shape, brand, pattern) of the identified item and searches the e-commerce site's database to obtain related product information.

[2004] Step 6:

[2005] The server retrieves relevant product information from the search results and ranks the most suitable products based on the viewer's past browsing and purchasing history, taking user preferences into account in the process.

[2006] Step 7:

[2007] The server runs an emotion engine that analyzes the viewer's facial and voice data to determine their emotions, in this case emotional states such as excitement, joy, and surprise.

[2008] Step 8:

[2009] The server further filters the ranked product information based on the user's emotional state, prioritizing products that best suit the user's current emotions.

[2010] Step 9:

[2011] The server sends the identified product information (product name, price, sales site link, image, etc.) to the device. This information is optimized based on emotions.

[2012] Step 10:

[2013] The product information received by the device is displayed on the screen in a format that is intuitive and easy for viewers to understand, including pop-ups and overlays.

[2014] Step 11:

[2015] Users review the displayed product information and click on the link to the product detail page of the product they are interested in. This action makes it easier to obtain additional information.

[2016] Step 12:

[2017] The device will redirect to an e-commerce site where the user can view detailed information, allowing them to make a final purchasing decision.

[2018] Example 2

[2019] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2020] Conventional video viewing systems lack a means for viewers to quickly and easily obtain information about items that interest them within video content. Furthermore, there is no method for providing information or encouraging purchasing behavior based on the viewer's emotional state. This makes it difficult to effectively stimulate viewers' purchasing desire and provide an optimal purchasing experience.

[2021] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for tapping an item in which a viewer is interested, a means for transmitting tapped position information and frame data to the server, a means for analyzing an object in the frame based on the transmitted information and identifying the relevant item, a means for searching a database based on characteristic information of the identified item and acquiring related product information, a means for analyzing facial expressions and voice data of the viewer and identifying the emotional state, and a means for filtering or ranking product information based on the identified emotional state. This makes it possible to quickly and easily acquire information about items in which the viewer is interested, and to optimally display information and promote purchases according to the viewer's emotions.

[2022] A "viewer" is a user who is viewing video content on a device.

[2023] A "device" is an electronic device (e.g., smartphone, tablet, computer) that displays video content and can be operated by a user.

[2024] "Video content" means digital media containing a combination of moving images and sounds that convey information through visual and auditory means.

[2025] An "item" refers to an object or person that appears in video content and is an object of interest to viewers.

[2026] "Tap" means that the viewer selects an item by touching the device's touchscreen.

[2027] "Tap location information" is the coordinate data when a viewer taps an item on the device screen.

[2028] "Frame data" refers to information about each frame (still image) of video content, and is data that represents video at a specific point in time.

[2029] A "server" is a central processing unit that receives data from devices and performs analytical processing and database searches.

[2030] "Object analysis means" refers to the technology or algorithms used by the server to identify objects or items present in each frame of video content (e.g., object recognition AI models).

[2031] The "means for identifying the relevant item" is a method for finding the item that the viewer tapped on from among the analyzed objects.

[2032] The "means for searching a database" is a method for querying a database in which various product information is stored based on the characteristic information of the identified item.

[2033] "Related Product Information" is detailed information (e.g., price, description, images, links) about products that are similar or match the identified item.

[2034] The "means for displaying to the viewer" refers to a method for visually presenting the acquired product information on a device.

[2035] "Means for analyzing facial and / or audio data" means the techniques or algorithms used to infer the emotional state of a viewer from their facial expressions or audio (e.g., emotion engine).

[2036] "Emotional state" is data that reflects the type and intensity of a viewer's emotion, including joy, excitement, sadness, etc.

[2037] A "filtering or ranking means" is a method for sorting or prioritizing retrieved product information based on the identified emotional state.

[2038] The present invention is a system that allows a viewer to quickly and easily obtain information about items that interest them while watching video content on a device, and provides information that corresponds to the viewer's emotions and promotes purchasing behavior.

[2039] The overall operation of the system involves three entities: the server, the device, and the viewer. We will explain how each entity interacts and processes information.

[2040] Server-side processing

[2041] First, we will explain the server-side processing. Specifically, it has the following functions:

[2042] 1. Video Analysis Module:

[2043] The server splits the video content received from the user into frames and identifies items within the frames using an object recognition AI model (e.g., YOLO or Detectron2). Through this analysis, the server identifies items that may be of interest to the viewer.

[2044] 2. Database Search:

[2045] Based on the analyzed item's attribute information (color, shape, brand logo, etc.), the server queries the e-commerce site's database (e.g., Amazon Product Advertising API, Rakuten product database) to obtain related product information.

[2046] 3. Ranking Algorithm:

[2047] Based on the acquired product information, the server takes into account the viewer's past browsing and purchasing history and uses AI algorithms (such as collaborative filtering or content-based filtering) to rank the products that are best suited to the viewer.

[2048] 4. Sentiment Analysis Module:

[2049] The server acquires the viewer's facial expression and voice data and analyzes it using an emotion engine (e.g., Microsoft Azure Emotion API, IBM Watson Tone Analyzer) to determine their real-time emotional state (excitement, joy, calmness, etc.).

[2050] 5. Emotion-based filtering:

[2051] The server filters or re-ranks product information based on the viewer's emotional state, for example, prioritizing special offers and limited edition products for an excited viewer.

[2052] 6. Provision of Information:

[2053] The server immediately transmits the filtered product information to the terminal and provides a link to the detailed page of the product the viewer needs.

[2054] Terminal side processing

[2055] Next, the processing on the terminal side will be described.

[2056] 1. Listening for tap events:

[2057] The device monitors the viewer's tapping on the screen while watching a video, acquires the location information of the tap, and sends this location information to the server.

[2058] 2. Data transmission:

[2059] The tapped position information and frame data are sent to the server, which uses this information in the server's video analysis module.

[2060] 3. Information Receipt and Display:

[2061] It receives product information sent from the server and displays related information on the screen while the user is watching. Specifically, it displays product information pop-ups, links to related e-commerce sites, thumbnail images of the products, and more.

[2062] User processing

[2063] Finally, the processing on the user side will be described.

[2064] 1. Item tap:

[2065] The user taps on an item of interest (e.g., the actress's clothing) in the video they are watching. This action records the location of the tap on the device.

[2066] 2. Information verification:

[2067] The user checks the product information displayed on the device, clicks on the link to the related e-commerce site, and views the product details page.

[2068] 3. Purchasing decision:

[2069] The user decides on a purchasing decision based on the information provided, and if they like the product, they complete the purchase process through the e-commerce site.

[2070] Specific examples

[2071] Next, a specific example of the operation of the present invention will be described.

[2072] 1. User Scenario:

[2073] While watching a drama starring an actress on his smartphone, a user becomes interested in the dress the actress is wearing and taps on the dress.

[2074] 2. Tap event occurs:

[2075] The device detects the tap event and sends the tap position and frame information to the server. Specifically, it sends the tap coordinates and the frame data at that time.

[2076] 3. Parsing on the server:

[2077] The server analyzes the items in the received frame and recognizes the dress. Based on the recognition results, it searches the e-commerce site database to identify the most suitable dress.

[2078] 4. Running the Emotion Engine:

[2079] The server analyzes the user's emotions from facial expressions and voice data, and determines, for example, whether the user is in an excited state.

[2080] 5. Emotional information:

[2081] The server provides information on products that suit the user's state of excitement with priority based on information from the emotion engine.

[2082] 6. Information provision:

[2083] The server sends detailed information about the dress and a link to purchase it to the device, which then displays the details of the dress to the user based on the information received.

[2084] 7. Purchasing behavior:

[2085] The user checks the displayed information, and if they like the dress, they click on the provided link to go to the e-commerce site and complete the purchase process.

[2086] Example prompt:

[2087] "If I want to know about the dress an actress is wearing, how can the server provide this information and explain the process for the user to purchase it?"

[2088] As described above, the system of the present invention is designed to quickly and easily obtain information about items that interest viewers, and to display information optimally in response to the viewers' emotions and promote purchases.

[2089] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2090] Step 1:

[2091] Listening for touch events

[2092] The device monitors the viewer's tapping on the screen while watching a video. The input is the viewer's tapping on the touchscreen. Specifically, the device obtains the tapped position information (coordinate data) and the frame data indicating which frame was being displayed. The output is the tapped position information and frame data.

[2093] Step 2:

[2094] Sending a request to the server

[2095] The device sends the acquired tap position information and frame data to the server. The inputs include tap position information and frame data. Specifically, the device sends this information to the server in the form of an HTTP request or similar. The output includes the tap position information and frame data sent to the server.

[2096] Step 3:

[2097] Running the video analysis module

[2098] The server uses the received frame data to run an object recognition AI model (e.g., YOLO or Detectron2) and analyze the objects in the frame. The input is the frame data. Specifically, the server analyzes the frame and identifies the item. The output is information about the analyzed item.

[2099] Step 4:

[2100] Database search of identification results

[2101] The server searches the e-commerce site's database based on the analyzed item's characteristic information (color, shape, brand logo, etc.). The analyzed item information is the input. Specifically, the server calls the API to retrieve related product information. The retrieved product information is the output.

[2102] Step 5:

[2103] Ranking of specific products

[2104] The server uses AI algorithms (such as collaborative filtering or content-based filtering) to rank the acquired product information. The inputs are the acquired product information and the viewer's past browsing history and purchase history. Specifically, the server uses AI to rank the products. The output is the ranked product information.

[2105] Step 6:

[2106] Running the Emotion Engine

[2107] The server acquires the viewer's facial expression and voice data and analyzes it using an emotion engine (e.g., Microsoft Azure Emotion API, IBM Watson Tone Analyzer). The input is the viewer's facial expression and voice data. Specifically, the server analyzes the emotion data and identifies the viewer's emotional state. The output is the analyzed emotional state.

[2108] Step 7:

[2109] Sentiment-based filtering

[2110] The server filters or re-ranks the product information based on the analyzed emotional state. The inputs include the analyzed emotional state and the ranked product information. As a specific operation, the server sorts or prioritizes the product information according to the emotional state. The output includes the filtered or re-ranked product information.

[2111] Step 8:

[2112] Providing information

[2113] The server transmits the filtered or re-ranked product information to the terminal. The input includes the filtered or re-ranked product information. As a specific operation, the server converts the product information into a data packet and transmits it to the terminal. The output includes the product information transmitted to the terminal.

[2114] Step 9:

[2115] Information reception and display

[2116] The device receives product information sent from the server and displays related information on the screen being viewed. The input is the product information sent from the server. In concrete terms, the device presents the product information to the user in the form of a pop-up display or related links. The output is the product information displayed in a format that the viewer can check.

[2117] Step 10:

[2118] Information verification and purchasing behavior

[2119] The user checks the product information displayed on the device and, if interested, proceeds with the purchase. The input is the product information displayed on the device. The specific operation is for the user to click on a link to the relevant e-commerce site and proceed with the purchase. The output is a result indicating whether the purchase was completed.

[2120] Through the above steps, viewers can easily obtain information about items that interest them, and information is displayed according to their emotions, promoting purchasing behavior.

[2121] (Application example 2)

[2122] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2123] Currently, there are insufficient ways for viewers to easily obtain information about items they are interested in while watching video content.In addition, information is not provided according to the viewer's emotional state, making it difficult to suggest optimal product information.

[2124] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[2125] In this invention, the server includes means for tapping on an object of interest to a viewer while watching video content on a device, means for transmitting tapped position information and frame data to the server, means for analyzing objects in the frame based on the transmitted information and identifying the relevant object, means for searching a database based on characteristic information of the identified object and acquiring related product information, means for transmitting the acquired product information to the device and displaying it to the viewer, means for analyzing the viewer's facial expressions and voice data using an emotion engine and identifying the viewer's emotional state, and means for filtering or ranking the product information to be provided based on the viewer's emotional state. This allows the viewer to quickly and easily obtain information on items that interest them while watching a video, and further makes it possible to provide optimal product information according to the viewer's emotional state.

[2126] A "viewer" is a person who watches video content.

[2127] A "device" is an electronic device that displays video content and allows viewers to interact with it.

[2128] "Video content" refers to all media content provided in video format.

[2129] An "object" is an object that a viewer becomes interested in while watching video content and tries to obtain information by tapping on it.

[2130] A "tap" is the act of touching a specific location on a device screen once with a finger or stylus.

[2131] "Frame data" refers to each still image data that constitutes video content.

[2132] A "server" is a computer system that receives a request over a network, executes the specified process, and returns a response.

[2133] "Analyzing an object" means recognizing a specific object from frame data in video content and extracting information about it.

[2134] "Feature information" refers to attribute information for identifying an object, such as the object's color, shape, brand logo, etc.

[2135] A "database" is a collection of digital information organized to allow efficient storage and retrieval of specific information.

[2136] "Product Information" refers to detailed information about products and services related to the subject matter.

[2137] An "emotion engine" is a software module that analyzes the viewer's emotional state and optimizes the information provided based on the results.

[2138] "Emotional state" refers to data that represents the viewer's current psychological and physiological state.

[2139] "Filtering" refers to the act of sorting and prioritizing the information provided.

[2140] "Ranking" is the act of assigning a rank to information or objects based on certain criteria.

[2141] The present invention is a system that easily obtains information about objects of interest to viewers while they are watching video content on their devices, and provides optimal product information according to the viewer's emotional state. The system for implementing the present invention consists of a server, a terminal, and user interactions.

[2142] Server Processing

[2143] The server consists of the following modules:

[2144] Object Recognition Module: Using TensorFlow and OpenCV, this module analyzes and identifies objects within frames of video content. This module recognizes objects based on the transmitted location information and frame data, and extracts their feature information.

[2145] Database search module: Using MySQL or MongoDB, this module searches the database for relevant product information based on the characteristics of the identified object. This module quickly retrieves and ranks product information.

[2146] Sentiment analysis engine: Using Google Cloud Vision and Amazon Rekognition, the engine analyzes the viewer's emotional state from facial expressions and voice data, and filters or ranks the product information provided based on the analysis results.

[2147] Terminal handling

[2148] The terminal has the following features:

[2149] Monitoring tap events: Events that occur when a viewer taps the screen of the device are monitored, and the location information and frame data are sent to the server.

[2150] Communication with the server: When a tap event is detected, the information is sent to the server as a request and product information sent from the server is received.

[2151] Information display: Product information obtained from the server is displayed on the screen while the content is being viewed, and related product links and thumbnail images are provided.

[2152] User Action

[2153] The user interacts with:

[2154] Item Tap: Tap on an object that interests you while watching and send its location information.

[2155] Check information: Check the product information displayed on your device and click on the link for more information or to purchase.

[2156] Purchasing behavior: Select a product based on the information provided and complete the purchase process on the e-commerce site.

[2157] Specific examples

[2158] For example, suppose a user is watching a movie on a media distribution platform and is interested in the clothing worn by the lead actor. In this case, the user taps on the device screen. This tap event is detected by the device, and the relevant location information and frame data are sent to the server. The server identifies the clothing based on this data and searches for related product information from a database based on its feature information. Furthermore, an emotion analysis engine analyzes the user's emotional state (e.g., excitement or curiosity), ranks the product information accordingly, and sends it to the device. The device displays the results to the user, who can click on the provided link to check more information and make a purchase.

[2159] Prompt Sentence Examples

[2160] While a user is watching a movie on their smartphone, they tap on an item they are interested in. Can you give us an example of a system that retrieves information about that item and provides relevant product information, taking into account the user's emotional state?

[2161] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2162] Step 1:

[2163] User interaction and tap events

[2164] While a user is watching video content on a device, they tap on an object that interests them.

[2165] Input: User tap position information and video frame data

[2166] What happens: The device's touchscreen detects the user's taps and captures their location.

[2167] Step 2:

[2168] Sending tap position

[2169] The terminal transmits the tap position information and the associated frame data to the server.

[2170] Input: User tap position information and video frame data

[2171] Output: Tap position information and frame data are sent to the server.

[2172] Specific operation: The device detects a touch event and sends the coordinates and current frame data to the server as an HTTP request.

[2173] Step 3:

[2174] Analysis of objects in the frame

[2175] The server analyzes the objects in the frame based on the transmitted location information and frame data, and identifies the relevant target.

[2176] Input: Tap position information and video frame data

[2177] Output: Feature information of identified objects

[2178] Specific operation: The object recognition module (TensorFlow or OpenCV) on the server recognizes the object around the tap position in the frame and extracts its feature information.

[2179] Step 4:

[2180] Database search

[2181] The server searches the database based on the characteristic information of the identified object and obtains related product information.

[2182] Input: Feature information of the identified object

[2183] Output: List of product details

[2184] Specific operation: Using the feature information, query the database (MySQL or MongoDB) to retrieve the corresponding product information.

[2185] Step 5:

[2186] Emotional state analysis

[2187] The server uses an emotion analysis engine to identify the viewer's emotional state based on the user's facial expression data and voice data.

[2188] Input: User's facial expression data and voice data

[2189] Output: Emotional state assessment result

[2190] Specific operation: Analyzes the user's emotional state using APIs such as Google Cloud Vision and Amazon Rekognition.

[2191] Step 6:

[2192] Filtering and ranking product information

[2193] The server filters or ranks the product information based on the analyzed emotional state.

[2194] Input: List of product information, emotional state evaluation results

[2195] Output: Filtered or ranked product information

[2196] Specific behavior: Select and prioritize the most relevant product information based on emotional state.

[2197] Step 7:

[2198] Sending and Displaying Information

[2199] The server transmits the filtered product information to the terminal, which displays it on the screen being viewed.

[2200] Input: Filtered or ranked product information

[2201] Output: Display information on the user's device (product image, link to details, etc.)

[2202] Specific operation: The server sends the generated product information back to the device as an HTTP response, and the device receives the information and displays it on the screen.

[2203] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2204] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2205] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2206] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2207] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2208] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2209] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2210] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2211] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2212] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2213] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2214] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2215] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[2216] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[2217] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[2218] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[2219] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[2220] As an example of a system configured with a single processor, first, one processor is configured by ...

Claims

1. A way for viewers to tap on items of interest while watching video content on their device; means for transmitting tapped position information and frame data to a server; A means for analyzing an object in the frame based on the transmitted information and identifying the corresponding item; A means for searching a database based on the characteristic information of the identified item and acquiring related product information; means for transmitting the obtained product information to a device and displaying the information to a viewer; A system including:

2. A means for ranking related product information based on a viewer's past browsing history and purchasing history; A means for providing viewers with optimal product information based on the ranked information; The system of claim 1 further comprising:

3. A means for receiving an interaction from a viewer to view detailed information from the provided product information; A way to transition to an e-commerce site to display more information, The system of claim 1 further comprising:

Images