System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving a moving image file from a user terminal and storing the moving image file in a database; means for analyzing contents of the moving image and specifying a scene or a topic; means for selecting a related advertisement material based on the specified scene or topic; means for automatically generating an advertisement moving image by combining the selected advertisement materials; means for naturally connecting the generated advertisement moving image to the original moving image; and means for distributing the modified moving image to the user terminal.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 the modern video advertising market, large amounts of money are being invested in creating advertisements to attract viewer attention, but traditional advertising methods have limited effectiveness and end up increasing viewer stress. Furthermore, if advertisements are not properly integrated, viewers tend to avoid them, resulting in a decline in the return on investment for advertising companies. Furthermore, there is also the issue that direct revenue from ad viewing is not returned to content creators, preventing improvements in the quality and quantity of content. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing a system including: means for receiving video files from a user terminal and storing them in a database; means for analyzing the content of the video and identifying scenes and topics; means for selecting relevant advertising materials based on the identified scenes and topics; means for automatically generating an advertising video by combining the selected advertising materials; means for seamlessly stitching the generated advertising video to the original video; means for delivering the modified video to the user terminal; means for tracking the viewing status of the modified video and aggregating viewing data; and means for providing compensation to content creators based on the viewing data. This system reduces viewer stress, reduces corporate advertising costs, and improves advertising effectiveness. It also allows content creators to receive compensation directly, promoting improvements in the quality and quantity of content.

[0006] A "user terminal" is an electronic device used by an end user to upload and view videos.

[0007] A "video file" is a digital file that contains video and audio data.

[0008] A "database" is a system for storing and managing digital information such as video files and advertising materials, and the collection of data contained therein.

[0009] "Analysis" is the process of analyzing the video and audio data of a video to identify scenes and topics.

[0010] A "scene" is a part of a group of consecutive frames in a video that shows a specific content or situation.

[0011] "Topics" are subjects or topics extracted based on the audio data within the video.

[0012] "Advertising materials" are data such as images, videos, and text provided by companies for promotional purposes.

[0013] "Automatic generation" is the process of automatically creating an advertising video using a program based on selected advertising materials.

[0014] "Naturally splicing" means editing scenes so that the altered video does not disrupt the flow of the original video.

[0015] "Distribution" is the process of transmitting the modified video to a user terminal over a network.

[0016] The "viewing status" is data indicating the status of a video, such as the start of playback, the end of playback, or skipping, when the user plays the video.

[0017] "Tracking" refers to the monitoring and recording of user behavior and viewing habits.

[0018] "Aggregation" refers to the statistical compilation of collected viewing data.

[0019] "Remuneration" refers to compensation such as money or points provided to content creators based on viewing data. [Brief explanation of the drawings]

[0020] [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

[0021] 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.

[0022] First, the terms used in the following description will be explained.

[0023] 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).

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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."

[0028] [First embodiment]

[0029] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0030] 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.

[0031] 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).

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 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.

[0038] 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.

[0039] 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.

[0040] 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."

[0041] System Overview

[0042] This invention is a system that uses AI to automatically generate advertisements appropriate for the situation for videos provided by users, and provides advertisements to viewers in a natural way. This system is composed of a user terminal, a server, and a database.

[0043] Specific operation of the system

[0044] Video upload and analysis

[0045] Users upload video files from their devices through a dedicated website or application. The server receives the video files and stores them in a database. The server then breaks down the stored video into frames and extracts the image data for each frame. The server uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.), and then uses speech recognition algorithms to convert the audio data into text. This allows the server to identify the scenes and topics in the video.

[0046] Selecting advertising materials and generating advertising videos

[0047] Based on the identified scenes and topics, the server selects relevant advertising materials from the advertising provider's database. Selection criteria include the relevance to the scene or topic, and the preferences of the target audience. The server then combines the selected advertising materials and automatically generates an advertising video using a video editing algorithm. The generated advertising video is edited to seamlessly integrate with the flow of the original video.

[0048] Video streaming and viewing

[0049] The modified video file is then delivered to the user's device by the server. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video. The user then plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements.

[0050] Aggregating viewing data and providing rewards

[0051] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing data. This allows content creators to receive rewards based on the viewing data. The server then credits these rewards to the content creators' accounts as electronic money or points.

[0052] Specific examples

[0053] For example, consider the case where a user uploads a travel vlog video. The server analyzes the video frame by frame and identifies scenes such as "introductory tourist spots" and "tasting local cuisine." Next, it selects advertising materials for sightseeing tours related to the identified "introduction of tourist spots" scene and automatically generates a 15-second advertising video. The generated advertising video is spliced ​​in immediately after the "introduction of tourist spots" scene, so that it is displayed in a natural flow for the viewer.

[0054] When viewers play the modified videos, the server tracks viewing data, which is then used to provide compensation to content creators, enabling them to generate direct revenue and businesses to more effectively deliver advertising.

[0055] In this way, the present invention is a system that simultaneously reduces corporate advertising costs, improves advertising effectiveness, reduces viewer stress, and provides rewards to content creators.

[0056] The processing flow will be explained below.

[0057] Step 1:

[0058] Users upload video files from their devices via a dedicated website or application, and the server receives the video files and stores them in a database.

[0059] Step 2:

[0060] The server breaks down the stored video into frames and extracts the image data for each frame.The server then uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.).

[0061] Step 3:

[0062] The server extracts audio data from the video, converts it into text using a speech recognition algorithm, and then analyzes the text data to identify topics and scenes.

[0063] Step 4:

[0064] The server selects relevant advertising materials from the database of advertising providers based on the identified scenes and topics, taking into account the relevance to the scenes and topics and the preferences of the target audience as selection criteria.

[0065] Step 5:

[0066] The server uses the selected advertising materials and a video editing algorithm to automatically generate a 15-second advertising video.The server then edits the generated advertising video so that it fits naturally into the original video without disrupting the flow of the original video.

[0067] Step 6:

[0068] The server adds metadata (advertising information, scene information, etc.) to the modified video file and provides a link and download option for distribution to the user's device.

[0069] Step 7:

[0070] The user plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the ad playback status (start, stop, skip, etc.).

[0071] Step 8:

[0072] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing data.The server then processes the content creators to provide rewards based on the viewing data.The rewards are credited to the content creators' accounts as electronic money or points.

[0073] Step 9:

[0074] The server generates detailed reports on ad viewing data and effectiveness for ad providers, allowing them to analyze the cost-effectiveness of advertising and provide feedback for future advertising strategies and optimization.

[0075] Example 1

[0076] 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."

[0077] Conventional ad distribution systems have struggled to insert ads into video content at appropriate and natural timing. Furthermore, many ads are intrusive to viewers, and compensation mechanisms for content creators are often unclear. This leads to lower viewer satisfaction and insufficient advertising effectiveness. Furthermore, there is a need for accurate tracking of viewing data and the provision of appropriate compensation to content creators.

[0078] 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.

[0079] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for breaking down the video into frames and extracting image data for each frame, means for analyzing the extracted image data to identify content such as people, objects, and backgrounds, means for converting audio data into text to identify scenes and topics, means for selecting relevant advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for naturally splicing the generated advertising video into the original video, and means for delivering the modified video to the user terminal. This makes it possible to insert advertisements into video content in a natural way and increase viewer satisfaction. It also makes it possible to accurately track viewing data and provide compensation to content creators based on the viewing data.

[0080] "User terminal" refers to a device used by a user, such as a computer, smartphone, or tablet.

[0081] A "video file" refers to a digital file format in which video and audio are recorded.

[0082] A "database" refers to a system for efficiently managing and storing data.

[0083] A "frame" refers to an individual still image that makes up a video.

[0084] "Image data" refers to digitized data of the visual information contained in each frame.

[0085] "Image recognition algorithms" refer to techniques that allow computers to understand objects and scenes in images.

[0086] A "voice recognition algorithm" refers to the technology that analyzes voice data and converts it into text.

[0087] A "scene" is a unit that represents a specific situation or situation within a video.

[0088] "Topics" refer to the themes or topics touched upon in the video.

[0089] "Advertising Materials" means information or content used for advertising.

[0090] "Video editing algorithm" refers to technology that automatically performs editing such as joining videos, trimming, and adding effects.

[0091] "Altered videos" refer to videos that have been edited by inserting advertisements into the original video.

[0092] "Metadata" refers to data about data, i.e., additional information about the original data.

[0093] "Tracking" refers to a system tracking user actions and data.

[0094] "Viewing data" refers to information about a user's viewing of a video (such as playback time, whether or not the video was skipped, and whether or not the ad was viewed).

[0095] "Reward" refers to the compensation provided by the system to content creators.

[0096] "Content creator" refers to an individual or organization that produces digital content such as videos.

[0097] This invention is a system that automatically generates advertisements appropriate for the situation for videos provided by users and provides advertisements to viewers in a natural way. This system is composed of a user terminal, a server, and a database.

[0098] System Overview

[0099] The server receives video files from user devices and stores them in a database. The server then breaks down the video into frames and extracts the image data for each frame. Software used includes OpenCV and TensorFlow. This allows the server to analyze the image data for each frame and identify content such as people, objects, and background. Google Cloud Speech-to-Text can also be used to convert audio data contained in the video into text and identify scenes and topics.

[0100] Based on the identified scenes and topics, the server selects relevant advertising materials from a database. The software used for this purpose includes, for example, a link to the advertising provider's database and custom scripts to implement the selection algorithm. The server then combines the selected advertising materials and automatically generates an advertising video using FFmpeg or Adobe Premiere Pro APIs.

[0101] The generated advertising video is then seamlessly integrated into the original video and delivered to the user's device. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video file.

[0102] The user plays the modified video on their device and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements. Based on the collected viewing data, the server compiles ad viewing rates and effectiveness indicators, and provides rewards to content creators based on the viewing data. These rewards are credited to the content creator's account as electronic money or points.

[0103] Specific examples

[0104] For example, consider the case where a user uploads a travel vlog video. The server analyzes the video frame by frame and identifies scenes such as "introductory scenes to tourist attractions" and "tasting local cuisine." Next, it selects advertising materials for sightseeing tours related to the identified "introduction scenes" and automatically generates a 15-second advertising video. This advertising video is spliced ​​in immediately after the "introduction scenes to tourist attractions" so that it appears natural to the viewer.

[0105] When viewers play the modified videos, the server tracks viewing data, which is then used to provide compensation to content creators, enabling them to generate direct revenue and businesses to more effectively deliver advertising.

[0106] Prompt Sentence Examples

[0107] "Upload a travel vlog video, analyze it, and automatically generate an ad video related to that scene. For example, insert an ad for a sightseeing tour into a scene introducing a tourist spot, and edit it so that it is presented to viewers in a natural flow."

[0108] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0109] Step 1:

[0110] Users upload video files from their devices using a dedicated website or application. At this time, users input the necessary metadata (video title, description, etc.) along with the video file. The input data, which consists of the video file and metadata, is then sent to the server.

[0111] Step 2:

[0112] The server receives the video file and metadata sent by the user and stores them in a database. The input data is the uploaded video file and metadata, and they are stored in the database.

[0113] Step 3:

[0114] The server uses FFmpeg to break down the stored video into frames and extract the image data for each frame. This process converts the input video file into a series of image data. The output data is the image data for each frame.

[0115] Step 4:

[0116] The server uses OpenCV and TensorFlow to analyze the image data of each frame. For example, it performs processing to identify people, objects, backgrounds, etc. The input data is the image data for each frame, and the output data is analyzed object information (for example, "people," "scenery," "buildings," etc.).

[0117] Step 5:

[0118] The server converts the audio data of the video into text using Google Cloud Speech-to-Text. Through this process, the input data is the audio data in the video, and the output data is the converted text data.

[0119] Step 6:

[0120] The server identifies scenes and topics based on the results of analyzing the image data and audio data. For example, it identifies scenes such as "introductions to tourist spots" or "tasting local cuisine." The input data are the results of analyzing the image data and audio data, and the output data is information about the identified scenes and topics.

[0121] Step 7:

[0122] The server selects relevant advertising materials from a database based on the identified scene and topic. This selection takes into account the relevance to the scene or topic and the preferences of the target audience. The input data is information about the scene or topic, and the output data is the selected advertising materials.

[0123] Step 8:

[0124] The server combines the selected advertising materials and automatically generates an advertising video using FFmpeg and Adobe Premiere Pro API. For example, it edits the video so that an advertisement for a sightseeing tour seamlessly follows a scene introducing a tourist spot. The input data is the advertising materials, and the output data is the generated advertising video.

[0125] Step 9:

[0126] The server creates a modified video by seamlessly splicing the generated advertising video into the original video. In this process, the input data is the original video and the generated advertising video, and the output data is the modified video.

[0127] Step 10:

[0128] The server delivers the modified video file to the user's device, and at the same time adds metadata (advertising information, scene information, etc.) to the modified video. The input data is the modified video and metadata, and the output data is the modified video to be delivered.

[0129] Step 11:

[0130] The user starts playing the modified video on the device. The server tracks the user's viewing behavior in real time and records the playback status of the advertisement. The input data is the user's viewing behavior, and the output data is the viewing data.

[0131] Step 12:

[0132] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing data, and provides rewards to content creators based on the viewing data. The input data is the viewing data, and the output data is the compilation results and rewards. Rewards are credited to the content creator's account as electronic money or points.

[0133] (Application example 1)

[0134] 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."

[0135] With conventional video advertising systems, the content of ads often did not match the video scenes or topics, causing viewers to feel uncomfortable. Furthermore, they were unable to analyze viewing data to maximize advertising effectiveness and real-time ad repositioning, which resulted in insufficient advertising effectiveness. Furthermore, they were also unable to provide appropriate rewards based on viewing data.

[0136] 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.

[0137] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for analyzing the content of the video and identifying scenes and topics, means for selecting relevant advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for seamlessly connecting the generated advertising video to the original video, means for delivering the modified video to the user terminal, means for tracking viewing data in real time, and means for rearranging the advertising video based on the viewing data. This makes it possible to insert appropriate advertisements without causing viewers a sense of incongruity and maximize advertising effectiveness based on the viewing data. It is also possible to provide appropriate rewards.

[0138] A "user terminal" is an electronic device that a user uses to shoot or upload videos.

[0139] A "database" is a system that systematically stores and manages data such as video files and advertising materials.

[0140] "Means for analyzing the content of video" refers to a processing device or software that breaks down a video file into frames and identifies scenes and topics using image and voice recognition.

[0141] "Means for identifying scenes and topics" refers to technology that analyzes image data and audio data to identify specific scenes and topics within a video.

[0142] The "means for selecting advertising material" is a technique for selecting relevant advertising content from an advertising database based on a specified scene or topic.

[0143] The "means for automatically generating advertising videos" is a video editing algorithm that naturally connects selected advertising materials to the original video.

[0144] The "means for distributing modified videos" refers to a technology that transmits video files with inserted advertisements to users' terminals.

[0145] "Means for tracking viewing data in real time" refers to technology that records the actions of users when they view modified videos in real time.

[0146] The "means for rearranging advertising videos" is a technology that rearranges advertisements to optimal positions based on viewing data.

[0147] A "reward mechanism" is a system or algorithm for rewarding content creators based on aggregated viewing data.

[0148] This invention is a system that uses AI to automatically generate advertisements appropriate for the situation for videos provided by users, and provides advertisements to viewers in a natural way. This system is composed of a user terminal, a server, and a database.

[0149] First, a video file is uploaded from the user's device to the server. The server then stores this video file in a database. The server then breaks down the stored video into frames and extracts image and audio data. The server then uses an image recognition algorithm to analyze the content of each frame (people, objects, background, etc.), and uses a speech recognition algorithm to convert the audio data into text. This analysis can be performed using Google Cloud's Speech-to-Text API or an open-source image recognition library.

[0150] Next, the server selects relevant advertising materials from an advertising database based on the identified scenes and topics. Selection criteria include relevance to the scene or topic, and the preferences of the target audience. The server combines the selected advertising materials and automatically generates an advertising video using libraries such as MoviePy. The generated advertising video is edited to seamlessly integrate with the flow of the original video.

[0151] The modified video file is delivered to the user's terminal by the server. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video. The user plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements. The viewing data is periodically compiled, and advertising viewing rates and effectiveness indicators are calculated. Based on this viewing data, the server can also reposition the advertising videos.

[0152] Furthermore, the server provides rewards to content creators based on the collected viewing data, which are credited to their accounts as electronic money or points.

[0153] For example, if a user uploads a "travel vlog video," the server analyzes the video frame by frame to identify scenes such as "introductory tourist spots" and "tasting local cuisine." The server then selects advertising materials for sightseeing tours related to the identified "introduction of tourist spots" scene and automatically generates a 15-second advertising video. The generated advertising video is spliced ​​in immediately after the "introduction of tourist spots" scene, creating a natural flow for viewers. When viewers play this modified video, the server tracks viewing data. Content creators are then rewarded based on the viewing data.

[0154] Example prompt for a generative AI model:

[0155] Perform scene analysis on user-provided videos, generate and insert appropriate ads. User video: Travel Vlog Scene 1: Tourist attraction introduction Scene 2: Tasting local cuisine Select ads related to the scene from the ad database, and edit the video to make it look natural.

[0156] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0157] Step 1:

[0158] A user shoots or uploads a video through a smartphone app. The input is the video file uploaded by the user, and this video file is sent to the server. The server stores the received video file in a database. The output is the stored video file.

[0159] Step 2:

[0160] The server breaks down the stored video file into frames. The input is a video file, and the server converts this video into image data for each frame. The output is image data for each frame. This is done using libraries such as OpenCV.

[0161] Step 3:

[0162] The server uses an image recognition algorithm to analyze the image data for each frame and identify people, objects, backgrounds, etc. The input is the image data for each frame, and the output is tag information that indicates the content of each frame. Here, an open-source image recognition library is used.

[0163] Step 4:

[0164] The server uses a speech recognition algorithm to convert the audio data from the video into text. The input is the audio data from the video file, and the output is text data. This is done using Google Cloud's Speech-to-Text API.

[0165] Step 5:

[0166] The server identifies scenes and topics from image and audio data. The input is tag information and text data for each frame, and the output is the identified scene and topic information. A generative AI model is used to extract features from each scene and identify the topic.

[0167] Step 6:

[0168] The server selects relevant advertising materials based on the identified scenes and topics. The input is scene and topic information, and the output is advertising materials retrieved from an advertising database. Selection criteria take into account relevance to the scene or topic, preferences of the target audience, etc.

[0169] Step 7:

[0170] The server automatically generates an advertising video using the selected advertising materials. The input is the advertising materials and scene information, and the output is the generated advertising video. Video editing is performed using libraries such as MoviePy.

[0171] Step 8:

[0172] The server then seamlessly stitches the generated ad video back into the original video. The input is the original video and the ad video, and the output is the modified video. This process is also performed using libraries such as MoviePy.

[0173] Step 9:

[0174] The server delivers the modified video file to the user's device. The input is the modified video, and the output is the video sent to the user's device. At this time, metadata such as added advertising information and scene information is also delivered.

[0175] Step 10:

[0176] The server tracks the user's viewing behavior in real time when they play the modified video. The input is the user's viewing behavior data, and the output is the tracked viewing data. Real-time tracking is performed using sensors and log data.

[0177] Step 11:

[0178] The server repositions the ads based on the collected viewing data. The input is the tracked viewing data, and the output is the repositioned ad video.

[0179] Step 12:

[0180] The server provides rewards to content creators based on the aggregated viewing data. The input is viewing data and advertising effectiveness indicators, and the output is electronic money or points given as rewards.

[0181] 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.

[0182] System Overview

[0183] This invention is a system that uses AI to automatically generate advertisements appropriate for the situation for videos provided by users, recognizes users' emotions, and utilizes the data. This system is composed of a user terminal, a server, an emotion engine, and a database.

[0184] Specific operation of the system

[0185] Video upload and analysis

[0186] Users upload video files from their devices through a dedicated website or application. The server receives the video files and stores them in a database. The server then breaks down the stored video into frames and extracts the image data for each frame. The server uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.), and then uses speech recognition algorithms to convert the audio data into text. This allows the server to identify the scenes and topics in the video.

[0187] Selecting advertising materials and generating advertising videos

[0188] Based on the identified scenes and topics, the server selects relevant advertising materials from the advertising provider's database. Selection criteria include the relevance to the scene or topic, and the preferences of the target audience. The server then combines the selected advertising materials and automatically generates an advertising video using a video editing algorithm. The generated advertising video is edited to seamlessly integrate with the flow of the original video.

[0189] User emotion recognition and data utilization

[0190] While the video is playing, the server uses an emotion engine to analyze video and audio data obtained from the user's device in real time and collects user emotional data. This emotional data is measured based on the user's facial expressions, tone of voice, eye movements, etc. The server then analyzes the acquired emotional data and selects the most appropriate advertising material based on that data.

[0191] Video streaming and viewing

[0192] The modified video file is then delivered to the user's device by the server. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video. The user then plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements.

[0193] Aggregating viewing and emotion data and providing rewards

[0194] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing and emotional data. As a result, rewards based on the viewing and emotional data are provided to content creators. The server credits these rewards to the content creators' accounts as electronic money or points.

[0195] Specific examples

[0196] For example, consider the case where a user uploads a travel vlog video. The server analyzes this video frame by frame and identifies scenes such as "introductory tourist spots" and "tasting local cuisine." Next, it selects advertising materials for sightseeing tours related to the identified "introduction of tourist spots" scenes, and uses an emotion engine to select tour information that is likely to interest the user. It automatically generates a 15-second advertising video, optimizing it based on the user's emotional data.

[0197] When viewers play the modified videos, the server captures and tracks emotional data through the emotion engine. Content creators are then rewarded based on the viewing and emotional data, enabling direct revenue for content creators and enabling businesses to deliver effective ad delivery and emotional ad engagement.

[0198] In this way, the present invention is a system that simultaneously reduces corporate advertising costs, improves advertising effectiveness, reduces viewer stress, and provides rewards to content creators.

[0199] The processing flow will be explained below.

[0200] Step 1:

[0201] Users upload video files from their devices via a dedicated website or application, and the server receives the video files and stores them in a database.

[0202] Step 2:

[0203] The server breaks down the stored video into frames and extracts the image data for each frame.The server then uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.).

[0204] Step 3:

[0205] The server extracts audio data from the video, converts it into text using a speech recognition algorithm, and then analyzes the text data to identify scenes and topics.

[0206] Step 4:

[0207] The server selects relevant advertising materials from the database of advertising providers based on the identified scenes and topics, taking into account the relevance to the scenes and topics and the preferences of the target audience as selection criteria.

[0208] Step 5:

[0209] The server uses the selected advertising materials and a video editing algorithm to automatically generate a 15-second advertising video.The server then edits the generated advertising video so that it fits naturally into the original video without disrupting the flow of the original video.

[0210] Step 6:

[0211] While the video is playing, the server uses an emotion engine to analyze video and audio data obtained from the user's device in real time and collects emotional data from the user, which is measured based on facial expressions, tone of voice, eye movements, etc.

[0212] Step 7:

[0213] The server analyzes the acquired emotional data and selects the most suitable advertising material based on that data. If the emotional data indicates joy, it performs optimization processing such as selecting advertising material with more positive content.

[0214] Step 8:

[0215] The server adds metadata (advertising information, scene information, etc.) to the modified video file and provides a link and download option for distribution to the user's device.

[0216] Step 9:

[0217] The user plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the ad playback status (start, stop, skip, etc.).

[0218] Step 10:

[0219] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing and emotion data.The server then processes the content creators to receive rewards based on the viewing and emotion data.The rewards are credited to the content creators' accounts as electronic money or points.

[0220] Step 11:

[0221] The server generates detailed reports for advertising companies on the effectiveness of their ads based on viewing data and sentiment data, allowing them to analyze the cost-effectiveness of their ads and provide feedback for their next advertising strategy and optimization.

[0222] Example 2

[0223] 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."

[0224] Conventional advertising systems only select ads based on video scenes and topics, making it difficult to optimize ads according to user emotions. Furthermore, there were no systems that could collect and analyze user emotions in real time and reselect advertising materials based on that information. This limited the effectiveness of advertising and made it difficult to improve user engagement.

[0225] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0226] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for analyzing the content of the video and identifying scenes and topics, means for selecting related advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for naturally splicing the generated advertising video into the original video, means for delivering the modified video to the user terminal, means for collecting and analyzing user emotion data in real time, and means for reselecting advertising materials based on the user emotion data, thereby enabling advertising delivery optimized to the user's emotions.

[0227] A "user terminal" is a device used by a user to upload and play videos, and includes smartphones, tablets, personal computers, etc.

[0228] A "server" is a computer system that is responsible for receiving, storing, analyzing, generating and distributing advertising materials, and other processes for video files.

[0229] A "database" is a system for systematically storing and managing data such as video files, advertising materials, and analysis results.

[0230] A "video file" is a file containing video and audio data uploaded by a user.

[0231] "Analysis" is the process of breaking down video content frame by frame, identifying its content using image and voice recognition technology, and identifying scenes and topics.

[0232] A "scene" refers to a specific situation or situation within a video, and is a part of a video that consists of elements such as people, objects, and background.

[0233] "Topics" refer to the themes or topics covered in the video, such as introductions to tourist spots or product reviews.

[0234] "Advertising materials" refers to elements such as images, videos, and audio used for advertising distribution, and are promotional content provided by companies.

[0235] "Automatic generation" refers to the process by which a system programmatically generates advertising videos without manual intervention.

[0236] A "modified video" is a video file that has been edited by adding advertising material to the original video.

[0237] "Emotion data" refers to data relating to the user's emotional state extracted from facial expressions, tone of voice, eye movements, and the like.

[0238] "Collection" refers to the process of acquiring emotion data and viewing data from user terminals and storing them on a server.

[0239] "Analysis" is the process of identifying the user's emotional state based on collected emotional data and selecting advertising materials accordingly.

[0240] "Viewing data" refers to information about the user's actions when watching a video, including viewing time, number of views, clicks, and the like.

[0241] "Reward" refers to compensation such as electronic money or points provided to content creators based on viewing data and emotional data.

[0242] The present invention is a system that automatically generates advertisements appropriate for the situation for videos provided by users, recognizes the user's emotions, and utilizes the data. This system is composed of a user terminal, a server, an emotion engine, and a database.

[0243] First, a user uploads a video file through a dedicated website or application. The device used can be a smartphone, tablet, or personal computer. The video file is then received by a server and stored in a database. Specifically, the database may use SQL Server or MySQL.

[0244] The server then breaks down the stored video into frames and extracts the image data for each frame. It uses image recognition algorithms such as OpenCV and speech recognition algorithms such as the Google Speech-to-Text API. The server then uses these algorithms to identify scenes and topics in the video.

[0245] Based on the identified scenes and topics, the server selects relevant advertising materials from the advertising provider's database. Selection criteria include relevance to the scene or topic, and the preferences of the target audience. The server then combines the selected advertising materials and automatically generates an advertising video using a video editing algorithm (e.g., FFmpeg).

[0246] After generating the ad video, the server uses an emotion engine to collect video and audio data from the user's device in real time to obtain the user's emotional data. Microsoft Azure Cognitive Services is used to analyze the emotional data. By analyzing the collected emotional data, the server can understand the user's emotional state while watching and can reselect advertising materials based on that data.

[0247] The modified video file is then delivered to the user's device by the server. This delivery utilizes streaming technology, such as RTMP (Real-Time Message Protocol). Metadata is added to the modified video, and user viewing status is tracked. The user then plays the modified video on their device and begins watching.

[0248] Finally, the server compiles ad viewing rates and effectiveness indicators based on the collected viewing and emotion data. Based on the compilation results, rewards are provided to content creators using the viewing and emotion data. Rewards are credited to the content creators' accounts as electronic money or points.

[0249] As a concrete example, consider the case where a user uploads a travel vlog video. The server analyzes the video and identifies scenes such as "introducing tourist spots" and "tasting local cuisine." Next, it selects advertising materials related to these scenes and seamlessly stitches the automatically generated advertising videos into the original video. When a user plays this modified video, the server obtains emotional data in real time and delivers optimal advertising, thereby improving user engagement.

[0250] Example prompt: "Describe how an AI system can automatically generate relevant ads for travel vlogs that showcase tourist attractions when uploaded."

[0251] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0252] Step 1:

[0253] A user uploads a video.

[0254] Input: A video file selected by the user.

[0255] Output: The video file that will be sent to the server.

[0256] How it works: The user selects a video file through a dedicated website or application and clicks the upload button. The video file is then sent to the server via the Internet.

[0257] Step 2:

[0258] The server receives the video and stores it in a database.

[0259] Input: The video file submitted by the user.

[0260] Output: Video files stored in the database.

[0261] Specific operation: The server receives the video file and saves it in a database (e.g., MySQL or SQL Server). When saving, the video file's metadata (e.g., file name, upload date, etc.) is also registered.

[0262] Step 3:

[0263] The server analyzes the video.

[0264] Input: Video files stored in the database.

[0265] Output: Image data decomposed into frames, and audio data converted to text.

[0266] What it does: The server breaks down the video file into frames, extracts the image data for each frame using OpenCV, and converts the audio data to text using the Google Speech-to-Text API.

[0267] Step 4:

[0268] The server identifies the scene and topic.

[0269] Input: Image data decomposed frame by frame and audio data converted to text.

[0270] Output: Data about identified scenes and topics.

[0271] How it works: The server uses image recognition and text analysis algorithms to identify scenes (e.g., introductions to tourist attractions, tastings of local cuisine) and topics within the video.

[0272] Step 5:

[0273] The server selects the advertising material.

[0274] Input: Data about the identified scene and topic.

[0275] Output: Selected advertising material data.

[0276] Specific operation: Based on the analysis results, the server issues a query to the ad provider's database to obtain relevant ad material data. Selection criteria take into account the relevance to the scene or topic, the preferences of the target audience, etc.

[0277] Step 6:

[0278] The server generates the advertisement video.

[0279] Input: Selected advertising material data.

[0280] Output: Auto-generated ad video.

[0281] Specific operation: The server combines the advertising materials and automatically generates an advertising video using a video editing algorithm (e.g., FFmpeg). The generated advertising video is edited to seamlessly connect with the original video.

[0282] Step 7:

[0283] The server collects and analyzes the emotion data.

[0284] Input: Real-time video and audio data sent from the user terminal.

[0285] Output: Parsed emotion data.

[0286] How it works: The server uses an emotion engine (such as Microsoft Azure Cognitive Services) to collect and analyze the user's video and audio data in real time. The analysis results include the user's facial expressions, tone of voice, and eye movements.

[0287] Step 8:

[0288] The server delivers the video to the user's device.

[0289] Input: The modified video file.

[0290] Output: A streaming URL that can be played on the user's device.

[0291] Specific operation: The server uploads the modified video file to the streaming service, generates a streaming URL, and notifies the user of this URL.

[0292] Step 9:

[0293] The server collects viewing data and provides rewards.

[0294] Input: User viewing and emotion data.

[0295] Output: Aggregated advertising effectiveness metrics and rewards for content creators.

[0296] Specific operation: The server monitors video playback status and emotional data in real time, aggregates viewing data, and awards electronic money or points to content creators through a reward system based on the aggregated results.

[0297] (Application example 2)

[0298] 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."

[0299] Conventional advertising systems lack the ability to automatically select and generate effective advertising materials for video content provided by users, making it difficult to deliver ads that take into account user emotions and the content of scenes. Furthermore, technology for optimizing ads based on viewer emotions is immature, and compensation for content creators is limited to viewing data. The purpose of this invention is to solve these issues and maximize advertising effectiveness while providing excellent convenience for viewers and content creators.

[0300] 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.

[0301] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for analyzing the content of the video and identifying scenes and topics, means for selecting related advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for naturally splicing the generated advertising video into the original video, means for delivering the modified video to the user terminal, and means for analyzing user emotions in real time and optimizing the advertising materials using the data. This makes it possible to analyze user emotions and video scenes in real time and automatically generate and deliver optimized advertising videos.

[0302] A "user terminal" is a device used by a user, such as a computer, smartphone, or tablet.

[0303] A "video file" is a digital file containing video and audio provided by a user.

[0304] A "database" is a system for storing and managing video files and related information.

[0305] The "server" is a centralized management system for managing the database, analyzing videos, and generating advertisements.

[0306] A "scene" refers to a specific situation or situation within a video.

[0307] "Topics" are themes or topics covered in the video.

[0308] "Advertising materials" are content such as video, audio, and text that are the source of the advertisement to be distributed.

[0309] An "advertising video" is an advertising video inserted into a video provided by a user.

[0310] "Emotion" refers to the emotional response or state that a user shows when watching a video.

[0311] "Advertising optimization" is the act of selecting and editing optimal advertising materials based on user emotions and video scenes.

[0312] "Viewing data" is data that records the actions and reactions of users when they watch videos.

[0313] "Reward" refers to monetary or points provided to content creators based on viewing data and emotional data.

[0314] To implement this invention, several major pieces of hardware and software are required, such as a server, a user terminal, a database, and an emotion engine.

[0315] First, a user shoots a video file using their own device and uploads it to the server via a dedicated application or website. User devices can be PCs, smartphones, tablets, etc. The server receives the uploaded video file and stores it in a database. The database centrally manages video files and their metadata.

[0316] The server then breaks down the saved video file into frames and analyzes the image and audio data of each frame. This process utilizes image and audio recognition algorithms. Specifically, OpenCV is used to analyze the image frames, and DeepFace and other algorithms are used to recognize facial expressions. Natural language processing libraries (such as the Hugging Face transformer) are used for audio recognition.

[0317] The server analyzes the content of the video to identify its scenes and topics, then selects advertising materials appropriate for the scenes from a related database, taking into account the content of the scenes and the attributes of the target audience.

[0318] Once the ad material selection is complete, the server automatically generates an ad video using the selected ad material. At this time, a video editing library (e.g., MoviePy) is used to seamlessly stitch the ad video into the original video. The generated ad video is designed to maintain the flow of the original video.

[0319] The modified video is then sent back to the user's device from the server. The user watches the modified video, and during this process the server analyzes the user's emotions in real time. DeepFace and other emotion analysis engines are used as emotion engines. For example, the server collects user emotional data based on eye movements, facial expressions, and tone of voice.

[0320] Furthermore, the system tracks the viewing status of the modified videos and aggregates the viewing data. This viewing data and emotional data are used to provide compensation to content creators. The compensation is credited to the content creator's account in the form of electronic money or points.

[0321] For example, suppose a user uploads a travel vlog video. In this case, the server analyzes the video and identifies scenes such as "introducing tourist spots" and "tasting local cuisine." It then selects advertising materials for sightseeing tours that are suitable for the "introducing tourist spots" scene and automatically generates a 15-second advertising video based on them. The generated advertising video is then seamlessly integrated into the original video and optimized based on the user's emotional data.

[0322] An example of a prompt sentence is as follows:

[0323] Generate a Python program that performs frame-based emotion recognition and inserts appropriate ads.

[0324] By using such prompts, it is possible to have a generative AI model automatically generate a program to perform a specific task.

[0325] This makes it possible to analyze user emotions and video scenes in real time, and automatically generate and deliver optimized advertising videos.

[0326] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0327] Step 1:

[0328] Upload video files from the user's device

[0329] Users use their own devices to upload video files to the server through a dedicated application or website. At this time, the video files are stored in a database. The input is the video file recorded by the user, and the output is the video data stored in the database.

[0330] Step 2:

[0331] Video decomposition and frame analysis

[0332] The server breaks down the stored video file into frames. It uses OpenCV to analyze the image data of each frame and recognizes faces, objects, etc. The input is the video file stored in the database, and the output is the image data of each frame. Specifically, it uses an image recognition algorithm to analyze people, backgrounds, etc. in the video.

[0333] Step 3:

[0334] Audio analysis

[0335] The server uses a speech recognition algorithm to analyze the audio data in the video file, convert it into text, and identify the scenes and topics in the video. The input is the audio data for each frame, and the output is text data.

[0336] Step 4:

[0337] Identifying scenes and topics

[0338] The server integrates image data and audio data to identify scenes and topics within the video. Natural language processing algorithms and emotion recognition models are used here. The input is image data and audio text data for each frame, and the output is information on the identified scenes and topics.

[0339] Step 5:

[0340] Selection of advertising materials

[0341] The server selects the most suitable advertising material from a database based on the identified scene and topic, taking into account factors such as the relevance to the scene and the preferences of the target audience. The input is information about the scene and topic, and the output is the selected advertising material.

[0342] Step 6:

[0343] Ad video generation

[0344] The server automatically generates an ad video using the selected ad material. Using a video editing library such as MoviePy, the ad video is seamlessly spliced ​​into the original video. The input is the original video frames and ad material, and the output is the generated ad video.

[0345] Step 7:

[0346] Distribution of altered videos

[0347] The server integrates the generated advertising video with the original video and delivers the modified video file to the user terminal. The input is the generated advertising video and the original video, and the output is the modified video file.

[0348] Step 8:

[0349] User sentiment analysis

[0350] When a user watches the modified video, the server uses an emotion engine to analyze the user's reactions (facial expressions, gaze, tone of voice, etc.). The input is the video and audio data the user is watching, and the output is the user's emotional data.

[0351] Step 9:

[0352] Tracking and aggregating viewing data

[0353] The server tracks the viewing status of the modified video and aggregates the viewing data. The input is user viewing behavior data, and the output is aggregated viewing data.

[0354] Step 10:

[0355] Providing rewards

[0356] The server provides rewards to content creators based on the viewing and emotion data. The rewards are added to their accounts as electronic money or points. The input is the aggregated viewing and emotion data, and the output is the rewards to the content creators.

[0357] 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.

[0358] 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.

[0359] 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.

[0360] [Second embodiment]

[0361] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0362] 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.

[0363] 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).

[0364] 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.

[0365] 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.

[0366] 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).

[0367] 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. 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.

[0368] 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.

[0369] 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.

[0370] 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.

[0371] 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.

[0372] 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."

[0373] System Overview

[0374] This invention is a system that uses AI to automatically generate advertisements appropriate for the situation for videos provided by users, and provides advertisements to viewers in a natural way. This system is composed of a user terminal, a server, and a database.

[0375] Specific operation of the system

[0376] Video upload and analysis

[0377] Users upload video files from their devices through a dedicated website or application. The server receives the video files and stores them in a database. The server then breaks down the stored video into frames and extracts the image data for each frame. The server uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.), and then uses speech recognition algorithms to convert the audio data into text. This allows the server to identify the scenes and topics in the video.

[0378] Selecting advertising materials and generating advertising videos

[0379] Based on the identified scenes and topics, the server selects relevant advertising materials from the advertising provider's database. Selection criteria include the relevance to the scene or topic, and the preferences of the target audience. The server then combines the selected advertising materials and automatically generates an advertising video using a video editing algorithm. The generated advertising video is edited to seamlessly integrate with the flow of the original video.

[0380] Video streaming and viewing

[0381] The modified video file is then delivered to the user's device by the server. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video. The user then plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements.

[0382] Aggregating viewing data and providing rewards

[0383] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing data. This allows content creators to receive rewards based on the viewing data. The server then credits these rewards to the content creators' accounts as electronic money or points.

[0384] Specific examples

[0385] For example, consider the case where a user uploads a travel vlog video. The server analyzes the video frame by frame and identifies scenes such as "introductory tourist spots" and "tasting local cuisine." Next, it selects advertising materials for sightseeing tours related to the identified "introduction of tourist spots" scene and automatically generates a 15-second advertising video. The generated advertising video is spliced ​​in immediately after the "introduction of tourist spots" scene, so that it is displayed in a natural flow for the viewer.

[0386] When viewers play the modified videos, the server tracks viewing data, which is then used to provide compensation to content creators, enabling them to generate direct revenue and businesses to more effectively deliver advertising.

[0387] In this way, the present invention is a system that simultaneously reduces corporate advertising costs, improves advertising effectiveness, reduces viewer stress, and provides rewards to content creators.

[0388] The processing flow will be explained below.

[0389] Step 1:

[0390] Users upload video files from their devices via a dedicated website or application, and the server receives the video files and stores them in a database.

[0391] Step 2:

[0392] The server breaks down the stored video into frames and extracts the image data for each frame.The server then uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.).

[0393] Step 3:

[0394] The server extracts audio data from the video, converts it into text using a speech recognition algorithm, and then analyzes the text data to identify topics and scenes.

[0395] Step 4:

[0396] The server selects relevant advertising materials from the database of advertising providers based on the identified scenes and topics, taking into account the relevance to the scenes and topics and the preferences of the target audience as selection criteria.

[0397] Step 5:

[0398] The server uses the selected advertising materials and a video editing algorithm to automatically generate a 15-second advertising video.The server then edits the generated advertising video so that it fits naturally into the original video without disrupting the flow of the original video.

[0399] Step 6:

[0400] The server adds metadata (advertising information, scene information, etc.) to the modified video file and provides a link and download option for distribution to the user's device.

[0401] Step 7:

[0402] The user plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the ad playback status (start, stop, skip, etc.).

[0403] Step 8:

[0404] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing data.The server then processes the content creators to provide rewards based on the viewing data.The rewards are credited to the content creators' accounts as electronic money or points.

[0405] Step 9:

[0406] The server generates detailed reports on ad viewing data and effectiveness for ad providers, allowing them to analyze the cost-effectiveness of advertising and provide feedback for future advertising strategies and optimization.

[0407] Example 1

[0408] 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."

[0409] Conventional ad distribution systems have struggled to insert ads into video content at appropriate and natural timing. Furthermore, many ads are intrusive to viewers, and compensation mechanisms for content creators are often unclear. This leads to lower viewer satisfaction and insufficient advertising effectiveness. Furthermore, there is a need for accurate tracking of viewing data and the provision of appropriate compensation to content creators.

[0410] 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.

[0411] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for breaking down the video into frames and extracting image data for each frame, means for analyzing the extracted image data to identify content such as people, objects, and backgrounds, means for converting audio data into text to identify scenes and topics, means for selecting relevant advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for naturally splicing the generated advertising video into the original video, and means for delivering the modified video to the user terminal. This makes it possible to insert advertisements into video content in a natural way and increase viewer satisfaction. It also makes it possible to accurately track viewing data and provide compensation to content creators based on the viewing data.

[0412] "User terminal" refers to a device used by a user, such as a computer, smartphone, or tablet.

[0413] A "video file" refers to a digital file format in which video and audio are recorded.

[0414] A "database" refers to a system for efficiently managing and storing data.

[0415] A "frame" refers to an individual still image that makes up a video.

[0416] "Image data" refers to digitized data of the visual information contained in each frame.

[0417] "Image recognition algorithms" refer to techniques that allow computers to understand objects and scenes in images.

[0418] A "voice recognition algorithm" refers to the technology that analyzes voice data and converts it into text.

[0419] A "scene" is a unit that represents a specific situation or situation within a video.

[0420] "Topics" refer to the themes or topics touched upon in the video.

[0421] "Advertising Materials" means information or content used for advertising.

[0422] "Video editing algorithm" refers to technology that automatically performs editing such as joining videos, trimming, and adding effects.

[0423] "Altered videos" refer to videos that have been edited by inserting advertisements into the original video.

[0424] "Metadata" refers to data about data, i.e., additional information about the original data.

[0425] "Tracking" refers to a system tracking user actions and data.

[0426] "Viewing data" refers to information about a user's viewing of a video (such as playback time, whether or not the video was skipped, and whether or not the ad was viewed).

[0427] "Reward" refers to the compensation provided by the system to content creators.

[0428] "Content creator" refers to an individual or organization that produces digital content such as videos.

[0429] This invention is a system that automatically generates advertisements appropriate for the situation for videos provided by users and provides advertisements to viewers in a natural way. This system is composed of a user terminal, a server, and a database.

[0430] System Overview

[0431] The server receives video files from user devices and stores them in a database. The server then breaks down the video into frames and extracts the image data for each frame. Software used includes OpenCV and TensorFlow. This allows the server to analyze the image data for each frame and identify content such as people, objects, and background. Google Cloud Speech-to-Text can also be used to convert audio data contained in the video into text and identify scenes and topics.

[0432] Based on the identified scenes and topics, the server selects relevant advertising materials from a database. The software used for this purpose includes, for example, a link to the advertising provider's database and custom scripts to implement the selection algorithm. The server then combines the selected advertising materials and automatically generates an advertising video using FFmpeg or Adobe Premiere Pro APIs.

[0433] The generated advertising video is then seamlessly integrated into the original video and delivered to the user's device. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video file.

[0434] The user plays the modified video on their device and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements. Based on the collected viewing data, the server compiles ad viewing rates and effectiveness indicators, and provides rewards to content creators based on the viewing data. These rewards are credited to the content creator's account as electronic money or points.

[0435] Specific examples

[0436] For example, consider the case where a user uploads a travel vlog video. The server analyzes the video frame by frame and identifies scenes such as "introductory scenes to tourist attractions" and "tasting local cuisine." Next, it selects advertising materials for sightseeing tours related to the identified "introduction scenes" and automatically generates a 15-second advertising video. This advertising video is spliced ​​in immediately after the "introduction scenes to tourist attractions" so that it appears natural to the viewer.

[0437] When viewers play the modified videos, the server tracks viewing data, which is then used to provide compensation to content creators, enabling them to generate direct revenue and businesses to more effectively deliver advertising.

[0438] Prompt Sentence Examples

[0439] "Upload a travel vlog video, analyze it, and automatically generate an ad video related to that scene. For example, insert an ad for a sightseeing tour into a scene introducing a tourist spot, and edit it so that it is presented to viewers in a natural flow."

[0440] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0441] Step 1:

[0442] Users upload video files from their devices using a dedicated website or application. At this time, users input the necessary metadata (video title, description, etc.) along with the video file. The input data, which consists of the video file and metadata, is then sent to the server.

[0443] Step 2:

[0444] The server receives the video file and metadata sent by the user and stores them in a database. The input data is the uploaded video file and metadata, and they are stored in the database.

[0445] Step 3:

[0446] The server uses FFmpeg to break down the stored video into frames and extract the image data for each frame. This process converts the input video file into a series of image data. The output data is the image data for each frame.

[0447] Step 4:

[0448] The server uses OpenCV and TensorFlow to analyze the image data of each frame. For example, it performs processing to identify people, objects, backgrounds, etc. The input data is the image data for each frame, and the output data is analyzed object information (for example, "people," "scenery," "buildings," etc.).

[0449] Step 5:

[0450] The server converts the audio data of the video into text using Google Cloud Speech-to-Text. Through this process, the input data is the audio data in the video, and the output data is the converted text data.

[0451] Step 6:

[0452] The server identifies scenes and topics based on the results of analyzing the image data and audio data. For example, it identifies scenes such as "introductions to tourist spots" or "tasting local cuisine." The input data are the results of analyzing the image data and audio data, and the output data is information about the identified scenes and topics.

[0453] Step 7:

[0454] The server selects relevant advertising materials from a database based on the identified scene and topic. This selection takes into account the relevance to the scene or topic and the preferences of the target audience. The input data is information about the scene or topic, and the output data is the selected advertising materials.

[0455] Step 8:

[0456] The server combines the selected advertising materials and automatically generates an advertising video using FFmpeg and Adobe Premiere Pro API. For example, it edits the video so that an advertisement for a sightseeing tour seamlessly follows a scene introducing a tourist spot. The input data is the advertising materials, and the output data is the generated advertising video.

[0457] Step 9:

[0458] The server creates a modified video by seamlessly splicing the generated advertising video into the original video. In this process, the input data is the original video and the generated advertising video, and the output data is the modified video.

[0459] Step 10:

[0460] The server delivers the modified video file to the user's device, and at the same time adds metadata (advertising information, scene information, etc.) to the modified video. The input data is the modified video and metadata, and the output data is the modified video to be delivered.

[0461] Step 11:

[0462] The user starts playing the modified video on the device. The server tracks the user's viewing behavior in real time and records the playback status of the advertisement. The input data is the user's viewing behavior, and the output data is the viewing data.

[0463] Step 12:

[0464] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing data, and provides rewards to content creators based on the viewing data. The input data is the viewing data, and the output data is the compilation results and rewards. Rewards are credited to the content creator's account as electronic money or points.

[0465] (Application example 1)

[0466] 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."

[0467] With conventional video advertising systems, the content of ads often did not match the video scenes or topics, causing viewers to feel uncomfortable. Furthermore, they were unable to analyze viewing data to maximize advertising effectiveness and real-time ad repositioning, which resulted in insufficient advertising effectiveness. Furthermore, they were also unable to provide appropriate rewards based on viewing data.

[0468] 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.

[0469] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for analyzing the content of the video and identifying scenes and topics, means for selecting relevant advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for seamlessly connecting the generated advertising video to the original video, means for delivering the modified video to the user terminal, means for tracking viewing data in real time, and means for rearranging the advertising video based on the viewing data. This makes it possible to insert appropriate advertisements without causing viewers a sense of incongruity and maximize advertising effectiveness based on the viewing data. It is also possible to provide appropriate rewards.

[0470] A "user terminal" is an electronic device that a user uses to shoot or upload videos.

[0471] A "database" is a system that systematically stores and manages data such as video files and advertising materials.

[0472] "Means for analyzing the content of video" refers to a processing device or software that breaks down a video file into frames and identifies scenes and topics using image and voice recognition.

[0473] "Means for identifying scenes and topics" refers to technology that analyzes image data and audio data to identify specific scenes and topics within a video.

[0474] The "means for selecting advertising material" is a technique for selecting relevant advertising content from an advertising database based on a specified scene or topic.

[0475] The "means for automatically generating advertising videos" is a video editing algorithm that naturally connects selected advertising materials to the original video.

[0476] The "means for distributing modified videos" refers to a technology that transmits video files with inserted advertisements to users' terminals.

[0477] "Means for tracking viewing data in real time" refers to technology that records the actions of users when they view modified videos in real time.

[0478] The "means for rearranging advertising videos" is a technology that rearranges advertisements to optimal positions based on viewing data.

[0479] A "reward mechanism" is a system or algorithm for rewarding content creators based on aggregated viewing data.

[0480] This invention is a system that uses AI to automatically generate advertisements appropriate for the situation for videos provided by users, and provides advertisements to viewers in a natural way. This system is composed of a user terminal, a server, and a database.

[0481] First, a video file is uploaded from the user's device to the server. The server then stores this video file in a database. The server then breaks down the stored video into frames and extracts image and audio data. The server then uses an image recognition algorithm to analyze the content of each frame (people, objects, background, etc.), and uses a speech recognition algorithm to convert the audio data into text. This analysis can be performed using Google Cloud's Speech-to-Text API or an open-source image recognition library.

[0482] Next, the server selects relevant advertising materials from an advertising database based on the identified scenes and topics. Selection criteria include relevance to the scene or topic, and the preferences of the target audience. The server combines the selected advertising materials and automatically generates an advertising video using libraries such as MoviePy. The generated advertising video is edited to seamlessly integrate with the flow of the original video.

[0483] The modified video file is delivered to the user's terminal by the server. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video. The user plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements. The viewing data is periodically compiled, and advertising viewing rates and effectiveness indicators are calculated. Based on this viewing data, the server can also reposition the advertising videos.

[0484] Furthermore, the server provides rewards to content creators based on the collected viewing data, which are credited to their accounts as electronic money or points.

[0485] For example, if a user uploads a "travel vlog video," the server analyzes the video frame by frame to identify scenes such as "introductory tourist spots" and "tasting local cuisine." The server then selects advertising materials for sightseeing tours related to the identified "introduction of tourist spots" scene and automatically generates a 15-second advertising video. The generated advertising video is spliced ​​in immediately after the "introduction of tourist spots" scene, creating a natural flow for viewers. When viewers play this modified video, the server tracks viewing data. Content creators are then rewarded based on the viewing data.

[0486] Example prompt for a generative AI model:

[0487] Perform scene analysis on user-provided videos, generate and insert appropriate ads. User video: Travel Vlog Scene 1: Tourist attraction introduction Scene 2: Tasting local cuisine Select ads related to the scene from the ad database, and edit the video to make it look natural.

[0488] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0489] Step 1:

[0490] A user shoots or uploads a video through a smartphone app. The input is the video file uploaded by the user, and this video file is sent to the server. The server stores the received video file in a database. The output is the stored video file.

[0491] Step 2:

[0492] The server breaks down the stored video file into frames. The input is a video file, and the server converts this video into image data for each frame. The output is image data for each frame. This is done using libraries such as OpenCV.

[0493] Step 3:

[0494] The server uses an image recognition algorithm to analyze the image data for each frame and identify people, objects, backgrounds, etc. The input is the image data for each frame, and the output is tag information that indicates the content of each frame. Here, an open-source image recognition library is used.

[0495] Step 4:

[0496] The server uses a speech recognition algorithm to convert the audio data from the video into text. The input is the audio data from the video file, and the output is text data. This is done using Google Cloud's Speech-to-Text API.

[0497] Step 5:

[0498] The server identifies scenes and topics from image and audio data. The input is tag information and text data for each frame, and the output is the identified scene and topic information. A generative AI model is used to extract features from each scene and identify the topic.

[0499] Step 6:

[0500] The server selects relevant advertising materials based on the identified scenes and topics. The input is scene and topic information, and the output is advertising materials retrieved from an advertising database. Selection criteria take into account relevance to the scene or topic, preferences of the target audience, etc.

[0501] Step 7:

[0502] The server automatically generates an advertising video using the selected advertising materials. The input is the advertising materials and scene information, and the output is the generated advertising video. Video editing is performed using libraries such as MoviePy.

[0503] Step 8:

[0504] The server then seamlessly stitches the generated ad video back into the original video. The input is the original video and the ad video, and the output is the modified video. This process is also performed using libraries such as MoviePy.

[0505] Step 9:

[0506] The server delivers the modified video file to the user's device. The input is the modified video, and the output is the video sent to the user's device. At this time, metadata such as added advertising information and scene information is also delivered.

[0507] Step 10:

[0508] The server tracks the user's viewing behavior in real time when they play the modified video. The input is the user's viewing behavior data, and the output is the tracked viewing data. Real-time tracking is performed using sensors and log data.

[0509] Step 11:

[0510] The server repositions the ads based on the collected viewing data. The input is the tracked viewing data, and the output is the repositioned ad video.

[0511] Step 12:

[0512] The server provides rewards to content creators based on the aggregated viewing data. The input is viewing data and advertising effectiveness indicators, and the output is electronic money or points given as rewards.

[0513] 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.

[0514] System Overview

[0515] This invention is a system that uses AI to automatically generate advertisements appropriate for the situation for videos provided by users, recognizes users' emotions, and utilizes the data. This system is composed of a user terminal, a server, an emotion engine, and a database.

[0516] Specific operation of the system

[0517] Video upload and analysis

[0518] Users upload video files from their devices through a dedicated website or application. The server receives the video files and stores them in a database. The server then breaks down the stored video into frames and extracts the image data for each frame. The server uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.), and then uses speech recognition algorithms to convert the audio data into text. This allows the server to identify the scenes and topics in the video.

[0519] Selecting advertising materials and generating advertising videos

[0520] Based on the identified scenes and topics, the server selects relevant advertising materials from the advertising provider's database. Selection criteria include the relevance to the scene or topic, and the preferences of the target audience. The server then combines the selected advertising materials and automatically generates an advertising video using a video editing algorithm. The generated advertising video is edited to seamlessly integrate with the flow of the original video.

[0521] User emotion recognition and data utilization

[0522] While the video is playing, the server uses an emotion engine to analyze video and audio data obtained from the user's device in real time and collects user emotional data. This emotional data is measured based on the user's facial expressions, tone of voice, eye movements, etc. The server then analyzes the acquired emotional data and selects the most appropriate advertising material based on that data.

[0523] Video streaming and viewing

[0524] The modified video file is then delivered to the user's device by the server. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video. The user then plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements.

[0525] Aggregating viewing and emotion data and providing rewards

[0526] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing and emotional data. As a result, rewards based on the viewing and emotional data are provided to content creators. The server credits these rewards to the content creators' accounts as electronic money or points.

[0527] Specific examples

[0528] For example, consider the case where a user uploads a travel vlog video. The server analyzes this video frame by frame and identifies scenes such as "introductory tourist spots" and "tasting local cuisine." Next, it selects advertising materials for sightseeing tours related to the identified "introduction of tourist spots" scenes, and uses an emotion engine to select tour information that is likely to interest the user. It automatically generates a 15-second advertising video, optimizing it based on the user's emotional data.

[0529] When viewers play the modified videos, the server captures and tracks emotional data through the emotion engine. Content creators are then rewarded based on the viewing and emotional data, enabling direct revenue for content creators and enabling businesses to deliver effective ad delivery and emotional ad engagement.

[0530] In this way, the present invention is a system that simultaneously reduces corporate advertising costs, improves advertising effectiveness, reduces viewer stress, and provides rewards to content creators.

[0531] The processing flow will be explained below.

[0532] Step 1:

[0533] Users upload video files from their devices via a dedicated website or application, and the server receives the video files and stores them in a database.

[0534] Step 2:

[0535] The server breaks down the stored video into frames and extracts the image data for each frame.The server then uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.).

[0536] Step 3:

[0537] The server extracts audio data from the video, converts it into text using a speech recognition algorithm, and then analyzes the text data to identify scenes and topics.

[0538] Step 4:

[0539] The server selects relevant advertising materials from the database of advertising providers based on the identified scenes and topics, taking into account the relevance to the scenes and topics and the preferences of the target audience as selection criteria.

[0540] Step 5:

[0541] The server uses the selected advertising materials and a video editing algorithm to automatically generate a 15-second advertising video.The server then edits the generated advertising video so that it fits naturally into the original video without disrupting the flow of the original video.

[0542] Step 6:

[0543] While the video is playing, the server uses an emotion engine to analyze video and audio data obtained from the user's device in real time and collects emotional data from the user, which is measured based on facial expressions, tone of voice, eye movements, etc.

[0544] Step 7:

[0545] The server analyzes the acquired emotional data and selects the most suitable advertising material based on that data. If the emotional data indicates joy, it performs optimization processing such as selecting advertising material with more positive content.

[0546] Step 8:

[0547] The server adds metadata (advertising information, scene information, etc.) to the modified video file and provides a link and download option for distribution to the user's device.

[0548] Step 9:

[0549] The user plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the ad playback status (start, stop, skip, etc.).

[0550] Step 10:

[0551] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing and emotion data.The server then processes the content creators to receive rewards based on the viewing and emotion data.The rewards are credited to the content creators' accounts as electronic money or points.

[0552] Step 11:

[0553] The server generates detailed reports for advertising companies on the effectiveness of their ads based on viewing data and sentiment data, allowing them to analyze the cost-effectiveness of their ads and provide feedback for their next advertising strategy and optimization.

[0554] Example 2

[0555] 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."

[0556] Conventional advertising systems only select ads based on video scenes and topics, making it difficult to optimize ads according to user emotions. Furthermore, there were no systems that could collect and analyze user emotions in real time and reselect advertising materials based on that information. This limited the effectiveness of advertising and made it difficult to improve user engagement.

[0557] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0558] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for analyzing the content of the video and identifying scenes and topics, means for selecting related advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for naturally splicing the generated advertising video into the original video, means for delivering the modified video to the user terminal, means for collecting and analyzing user emotion data in real time, and means for reselecting advertising materials based on the user emotion data, thereby enabling advertising delivery optimized to the user's emotions.

[0559] A "user terminal" is a device used by a user to upload and play videos, and includes smartphones, tablets, personal computers, etc.

[0560] A "server" is a computer system that is responsible for receiving, storing, analyzing, generating and distributing advertising materials, and other processes for video files.

[0561] A "database" is a system for systematically storing and managing data such as video files, advertising materials, and analysis results.

[0562] A "video file" is a file containing video and audio data uploaded by a user.

[0563] "Analysis" is the process of breaking down video content frame by frame, identifying its content using image and voice recognition technology, and identifying scenes and topics.

[0564] A "scene" refers to a specific situation or situation within a video, and is a part of a video that consists of elements such as people, objects, and background.

[0565] "Topics" refer to the themes or topics covered in the video, such as introductions to tourist spots or product reviews.

[0566] "Advertising materials" refers to elements such as images, videos, and audio used for advertising distribution, and are promotional content provided by companies.

[0567] "Automatic generation" refers to the process by which a system programmatically generates advertising videos without manual intervention.

[0568] A "modified video" is a video file that has been edited by adding advertising material to the original video.

[0569] "Emotion data" refers to data relating to the user's emotional state extracted from facial expressions, tone of voice, eye movements, and the like.

[0570] "Collection" refers to the process of acquiring emotion data and viewing data from user terminals and storing them on a server.

[0571] "Analysis" is the process of identifying the user's emotional state based on collected emotional data and selecting advertising materials accordingly.

[0572] "Viewing data" refers to information about the user's actions when watching a video, including viewing time, number of views, clicks, and the like.

[0573] "Reward" refers to compensation such as electronic money or points provided to content creators based on viewing data and emotional data.

[0574] The present invention is a system that automatically generates advertisements appropriate for the situation for videos provided by users, recognizes the user's emotions, and utilizes the data. This system is composed of a user terminal, a server, an emotion engine, and a database.

[0575] First, a user uploads a video file through a dedicated website or application. The device used can be a smartphone, tablet, or personal computer. The video file is then received by a server and stored in a database. Specifically, the database may use SQL Server or MySQL.

[0576] The server then breaks down the stored video into frames and extracts the image data for each frame. It uses image recognition algorithms such as OpenCV and speech recognition algorithms such as the Google Speech-to-Text API. The server then uses these algorithms to identify scenes and topics in the video.

[0577] Based on the identified scenes and topics, the server selects relevant advertising materials from the advertising provider's database. Selection criteria include relevance to the scene or topic, and the preferences of the target audience. The server then combines the selected advertising materials and automatically generates an advertising video using a video editing algorithm (e.g., FFmpeg).

[0578] After generating the ad video, the server uses an emotion engine to collect video and audio data from the user's device in real time to obtain the user's emotional data. Microsoft Azure Cognitive Services is used to analyze the emotional data. By analyzing the collected emotional data, the server can understand the user's emotional state while watching and can reselect advertising materials based on that data.

[0579] The modified video file is then delivered to the user's device by the server. This delivery utilizes streaming technology, such as RTMP (Real-Time Message Protocol). Metadata is added to the modified video, and user viewing status is tracked. The user then plays the modified video on their device and begins watching.

[0580] Finally, the server compiles ad viewing rates and effectiveness indicators based on the collected viewing and emotion data. Based on the compilation results, rewards are provided to content creators using the viewing and emotion data. Rewards are credited to the content creators' accounts as electronic money or points.

[0581] As a concrete example, consider the case where a user uploads a travel vlog video. The server analyzes the video and identifies scenes such as "introducing tourist spots" and "tasting local cuisine." Next, it selects advertising materials related to these scenes and seamlessly stitches the automatically generated advertising videos into the original video. When a user plays this modified video, the server obtains emotional data in real time and delivers optimal advertising, thereby improving user engagement.

[0582] Example prompt: "Describe how an AI system can automatically generate relevant ads for travel vlogs that showcase tourist attractions when uploaded."

[0583] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0584] Step 1:

[0585] A user uploads a video.

[0586] Input: A video file selected by the user.

[0587] Output: The video file that will be sent to the server.

[0588] How it works: The user selects a video file through a dedicated website or application and clicks the upload button. The video file is then sent to the server via the Internet.

[0589] Step 2:

[0590] The server receives the video and stores it in a database.

[0591] Input: The video file submitted by the user.

[0592] Output: Video files stored in the database.

[0593] Specific operation: The server receives the video file and saves it in a database (e.g., MySQL or SQL Server). When saving, the video file's metadata (e.g., file name, upload date, etc.) is also registered.

[0594] Step 3:

[0595] The server analyzes the video.

[0596] Input: Video files stored in the database.

[0597] Output: Image data decomposed into frames, and audio data converted to text.

[0598] What it does: The server breaks down the video file into frames, extracts the image data for each frame using OpenCV, and converts the audio data to text using the Google Speech-to-Text API.

[0599] Step 4:

[0600] The server identifies the scene and topic.

[0601] Input: Image data decomposed frame by frame and audio data converted to text.

[0602] Output: Data about identified scenes and topics.

[0603] How it works: The server uses image recognition and text analysis algorithms to identify scenes (e.g., introductions to tourist attractions, tastings of local cuisine) and topics within the video.

[0604] Step 5:

[0605] The server selects the advertising material.

[0606] Input: Data about the identified scene and topic.

[0607] Output: Selected advertising material data.

[0608] Specific operation: Based on the analysis results, the server issues a query to the ad provider's database to obtain relevant ad material data. Selection criteria take into account the relevance to the scene or topic, the preferences of the target audience, etc.

[0609] Step 6:

[0610] The server generates the advertisement video.

[0611] Input: Selected advertising material data.

[0612] Output: Auto-generated ad video.

[0613] Specific operation: The server combines the advertising materials and automatically generates an advertising video using a video editing algorithm (e.g., FFmpeg). The generated advertising video is edited to seamlessly connect with the original video.

[0614] Step 7:

[0615] The server collects and analyzes the emotion data.

[0616] Input: Real-time video and audio data sent from the user terminal.

[0617] Output: Parsed emotion data.

[0618] How it works: The server uses an emotion engine (such as Microsoft Azure Cognitive Services) to collect and analyze the user's video and audio data in real time. The analysis results include the user's facial expressions, tone of voice, and eye movements.

[0619] Step 8:

[0620] The server delivers the video to the user's device.

[0621] Input: The modified video file.

[0622] Output: A streaming URL that can be played on the user's device.

[0623] Specific operation: The server uploads the modified video file to the streaming service, generates a streaming URL, and notifies the user of this URL.

[0624] Step 9:

[0625] The server collects viewing data and provides rewards.

[0626] Input: User viewing and emotion data.

[0627] Output: Aggregated advertising effectiveness metrics and rewards for content creators.

[0628] Specific operation: The server monitors video playback status and emotional data in real time, aggregates viewing data, and awards electronic money or points to content creators through a reward system based on the aggregated results.

[0629] (Application example 2)

[0630] 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."

[0631] Conventional advertising systems lack the ability to automatically select and generate effective advertising materials for video content provided by users, making it difficult to deliver ads that take into account user emotions and the content of scenes. Furthermore, technology for optimizing ads based on viewer emotions is immature, and compensation for content creators is limited to viewing data. The purpose of this invention is to solve these issues and maximize advertising effectiveness while providing excellent convenience for viewers and content creators.

[0632] 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.

[0633] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for analyzing the content of the video and identifying scenes and topics, means for selecting related advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for naturally splicing the generated advertising video into the original video, means for delivering the modified video to the user terminal, and means for analyzing user emotions in real time and optimizing the advertising materials using the data. This makes it possible to analyze user emotions and video scenes in real time and automatically generate and deliver optimized advertising videos.

[0634] A "user terminal" is a device used by a user, such as a computer, smartphone, or tablet.

[0635] A "video file" is a digital file containing video and audio provided by a user.

[0636] A "database" is a system for storing and managing video files and related information.

[0637] The "server" is a centralized management system for managing the database, analyzing videos, and generating advertisements.

[0638] A "scene" refers to a specific situation or situation within a video.

[0639] "Topics" are themes or topics covered in the video.

[0640] "Advertising materials" are content such as video, audio, and text that are the source of the advertisement to be distributed.

[0641] An "advertising video" is an advertising video inserted into a video provided by a user.

[0642] "Emotion" refers to the emotional response or state that a user shows when watching a video.

[0643] "Advertising optimization" is the act of selecting and editing optimal advertising materials based on user emotions and video scenes.

[0644] "Viewing data" is data that records the actions and reactions of users when they watch videos.

[0645] "Reward" refers to monetary or points provided to content creators based on viewing data and emotional data.

[0646] To implement this invention, several major pieces of hardware and software are required, such as a server, a user terminal, a database, and an emotion engine.

[0647] First, a user shoots a video file using their own device and uploads it to the server via a dedicated application or website. User devices can be PCs, smartphones, tablets, etc. The server receives the uploaded video file and stores it in a database. The database centrally manages video files and their metadata.

[0648] The server then breaks down the saved video file into frames and analyzes the image and audio data of each frame. This process utilizes image and audio recognition algorithms. Specifically, OpenCV is used to analyze the image frames, and DeepFace and other algorithms are used to recognize facial expressions. Natural language processing libraries (such as the Hugging Face transformer) are used for audio recognition.

[0649] The server analyzes the content of the video to identify its scenes and topics, then selects advertising materials appropriate for the scenes from a related database, taking into account the content of the scenes and the attributes of the target audience.

[0650] Once the ad material selection is complete, the server automatically generates an ad video using the selected ad material. At this time, a video editing library (e.g., MoviePy) is used to seamlessly stitch the ad video into the original video. The generated ad video is designed to maintain the flow of the original video.

[0651] The modified video is then sent back to the user's device from the server. The user watches the modified video, and during this process the server analyzes the user's emotions in real time. DeepFace and other emotion analysis engines are used as emotion engines. For example, the server collects user emotional data based on eye movements, facial expressions, and tone of voice.

[0652] Furthermore, the system tracks the viewing status of the modified videos and aggregates the viewing data. This viewing data and emotional data are used to provide compensation to content creators. The compensation is credited to the content creator's account in the form of electronic money or points.

[0653] For example, suppose a user uploads a travel vlog video. In this case, the server analyzes the video and identifies scenes such as "introducing tourist spots" and "tasting local cuisine." It then selects advertising materials for sightseeing tours that are suitable for the "introducing tourist spots" scene and automatically generates a 15-second advertising video based on them. The generated advertising video is then seamlessly integrated into the original video and optimized based on the user's emotional data.

[0654] An example of a prompt sentence is as follows:

[0655] Generate a Python program that performs frame-based emotion recognition and inserts appropriate ads.

[0656] By using such prompts, it is possible to have a generative AI model automatically generate a program to perform a specific task.

[0657] This makes it possible to analyze user emotions and video scenes in real time, and automatically generate and deliver optimized advertising videos.

[0658] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0659] Step 1:

[0660] Upload video files from the user's device

[0661] Users use their own devices to upload video files to the server through a dedicated application or website. At this time, the video files are stored in a database. The input is the video file recorded by the user, and the output is the video data stored in the database.

[0662] Step 2:

[0663] Video decomposition and frame analysis

[0664] The server breaks down the stored video file into frames. It uses OpenCV to analyze the image data of each frame and recognizes faces, objects, etc. The input is the video file stored in the database, and the output is the image data of each frame. Specifically, it uses an image recognition algorithm to analyze people, backgrounds, etc. in the video.

[0665] Step 3:

[0666] Audio analysis

[0667] The server uses a speech recognition algorithm to analyze the audio data in the video file, convert it into text, and identify the scenes and topics in the video. The input is the audio data for each frame, and the output is text data.

[0668] Step 4:

[0669] Identifying scenes and topics

[0670] The server integrates image data and audio data to identify scenes and topics within the video. Natural language processing algorithms and emotion recognition models are used here. The input is image data and audio text data for each frame, and the output is information on the identified scenes and topics.

[0671] Step 5:

[0672] Selection of advertising materials

[0673] The server selects the most suitable advertising material from a database based on the identified scene and topic, taking into account factors such as the relevance to the scene and the preferences of the target audience. The input is information about the scene and topic, and the output is the selected advertising material.

[0674] Step 6:

[0675] Ad video generation

[0676] The server automatically generates an ad video using the selected ad material. Using a video editing library such as MoviePy, the ad video is seamlessly spliced ​​into the original video. The input is the original video frames and ad material, and the output is the generated ad video.

[0677] Step 7:

[0678] Distribution of altered videos

[0679] The server integrates the generated advertising video with the original video and delivers the modified video file to the user terminal. The input is the generated advertising video and the original video, and the output is the modified video file.

[0680] Step 8:

[0681] User sentiment analysis

[0682] When a user watches the modified video, the server uses an emotion engine to analyze the user's reactions (facial expressions, gaze, tone of voice, etc.). The input is the video and audio data the user is watching, and the output is the user's emotional data.

[0683] Step 9:

[0684] Tracking and aggregating viewing data

[0685] The server tracks the viewing status of the modified video and aggregates the viewing data. The input is user viewing behavior data, and the output is aggregated viewing data.

[0686] Step 10:

[0687] Providing rewards

[0688] The server provides rewards to content creators based on the viewing and emotion data. The rewards are added to their accounts as electronic money or points. The input is the aggregated viewing and emotion data, and the output is the rewards to the content creators.

[0689] 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.

[0690] 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.

[0691] 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.

[0692] [Third embodiment]

[0693] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0694] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0695] 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).

[0696] 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.

[0697] 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.

[0698] 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).

[0699] 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. 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.

[0700] 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.

[0701] 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.

[0702] 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.

[0703] 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.

[0704] 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."

[0705] System Overview

[0706] This invention is a system that uses AI to automatically generate advertisements appropriate for the situation for videos provided by users, and provides advertisements to viewers in a natural way. This system is composed of a user terminal, a server, and a database.

[0707] Specific operation of the system

[0708] Video upload and analysis

[0709] Users upload video files from their devices through a dedicated website or application. The server receives the video files and stores them in a database. The server then breaks down the stored video into frames and extracts the image data for each frame. The server uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.), and then uses speech recognition algorithms to convert the audio data into text. This allows the server to identify the scenes and topics in the video.

[0710] Selecting advertising materials and generating advertising videos

[0711] Based on the identified scenes and topics, the server selects relevant advertising materials from the advertising provider's database. Selection criteria include the relevance to the scene or topic, and the preferences of the target audience. The server then combines the selected advertising materials and automatically generates an advertising video using a video editing algorithm. The generated advertising video is edited to seamlessly integrate with the flow of the original video.

[0712] Video streaming and viewing

[0713] The modified video file is then delivered to the user's device by the server. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video. The user then plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements.

[0714] Aggregating viewing data and providing rewards

[0715] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing data. This allows content creators to receive rewards based on the viewing data. The server then credits these rewards to the content creators' accounts as electronic money or points.

[0716] Specific examples

[0717] For example, consider the case where a user uploads a travel vlog video. The server analyzes the video frame by frame and identifies scenes such as "introductory tourist spots" and "tasting local cuisine." Next, it selects advertising materials for sightseeing tours related to the identified "introduction of tourist spots" scene and automatically generates a 15-second advertising video. The generated advertising video is spliced ​​in immediately after the "introduction of tourist spots" scene, so that it is displayed in a natural flow for the viewer.

[0718] When viewers play the modified videos, the server tracks viewing data, which is then used to provide compensation to content creators, enabling them to generate direct revenue and businesses to more effectively deliver advertising.

[0719] In this way, the present invention is a system that simultaneously reduces corporate advertising costs, improves advertising effectiveness, reduces viewer stress, and provides rewards to content creators.

[0720] The processing flow will be explained below.

[0721] Step 1:

[0722] Users upload video files from their devices via a dedicated website or application, and the server receives the video files and stores them in a database.

[0723] Step 2:

[0724] The server breaks down the stored video into frames and extracts the image data for each frame.The server then uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.).

[0725] Step 3:

[0726] The server extracts audio data from the video, converts it into text using a speech recognition algorithm, and then analyzes the text data to identify topics and scenes.

[0727] Step 4:

[0728] The server selects relevant advertising materials from the database of advertising providers based on the identified scenes and topics, taking into account the relevance to the scenes and topics and the preferences of the target audience as selection criteria.

[0729] Step 5:

[0730] The server uses the selected advertising materials and a video editing algorithm to automatically generate a 15-second advertising video.The server then edits the generated advertising video so that it fits naturally into the original video without disrupting the flow of the original video.

[0731] Step 6:

[0732] The server adds metadata (advertising information, scene information, etc.) to the modified video file and provides a link and download option for distribution to the user's device.

[0733] Step 7:

[0734] The user plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the ad playback status (start, stop, skip, etc.).

[0735] Step 8:

[0736] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing data.The server then processes the content creators to provide rewards based on the viewing data.The rewards are credited to the content creators' accounts as electronic money or points.

[0737] Step 9:

[0738] The server generates detailed reports on ad viewing data and effectiveness for ad providers, allowing them to analyze the cost-effectiveness of advertising and provide feedback for future advertising strategies and optimization.

[0739] Example 1

[0740] 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."

[0741] Conventional ad distribution systems have struggled to insert ads into video content at appropriate and natural timing. Furthermore, many ads are intrusive to viewers, and compensation mechanisms for content creators are often unclear. This leads to lower viewer satisfaction and insufficient advertising effectiveness. Furthermore, there is a need for accurate tracking of viewing data and the provision of appropriate compensation to content creators.

[0742] 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.

[0743] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for breaking down the video into frames and extracting image data for each frame, means for analyzing the extracted image data to identify content such as people, objects, and backgrounds, means for converting audio data into text to identify scenes and topics, means for selecting relevant advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for naturally splicing the generated advertising video into the original video, and means for delivering the modified video to the user terminal. This makes it possible to insert advertisements into video content in a natural way and increase viewer satisfaction. It also makes it possible to accurately track viewing data and provide compensation to content creators based on the viewing data.

[0744] "User terminal" refers to a device used by a user, such as a computer, smartphone, or tablet.

[0745] A "video file" refers to a digital file format in which video and audio are recorded.

[0746] A "database" refers to a system for efficiently managing and storing data.

[0747] A "frame" refers to an individual still image that makes up a video.

[0748] "Image data" refers to digitized data of the visual information contained in each frame.

[0749] "Image recognition algorithms" refer to techniques that allow computers to understand objects and scenes in images.

[0750] A "voice recognition algorithm" refers to the technology that analyzes voice data and converts it into text.

[0751] A "scene" is a unit that represents a specific situation or situation within a video.

[0752] "Topics" refer to the themes or topics touched upon in the video.

[0753] "Advertising Materials" means information or content used for advertising.

[0754] "Video editing algorithm" refers to technology that automatically performs editing such as joining videos, trimming, and adding effects.

[0755] "Altered videos" refer to videos that have been edited by inserting advertisements into the original video.

[0756] "Metadata" refers to data about data, i.e., additional information about the original data.

[0757] "Tracking" refers to a system tracking user actions and data.

[0758] "Viewing data" refers to information about a user's viewing of a video (such as playback time, whether or not the video was skipped, and whether or not the ad was viewed).

[0759] "Reward" refers to the compensation provided by the system to content creators.

[0760] "Content creator" refers to an individual or organization that produces digital content such as videos.

[0761] This invention is a system that automatically generates advertisements appropriate for the situation for videos provided by users and provides advertisements to viewers in a natural way. This system is composed of a user terminal, a server, and a database.

[0762] System Overview

[0763] The server receives video files from user devices and stores them in a database. The server then breaks down the video into frames and extracts the image data for each frame. Software used includes OpenCV and TensorFlow. This allows the server to analyze the image data for each frame and identify content such as people, objects, and background. Google Cloud Speech-to-Text can also be used to convert audio data contained in the video into text and identify scenes and topics.

[0764] Based on the identified scenes and topics, the server selects relevant advertising materials from a database. The software used for this purpose includes, for example, a link to the advertising provider's database and custom scripts to implement the selection algorithm. The server then combines the selected advertising materials and automatically generates an advertising video using FFmpeg or Adobe Premiere Pro APIs.

[0765] The generated advertising video is then seamlessly integrated into the original video and delivered to the user's device. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video file.

[0766] The user plays the modified video on their device and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements. Based on the collected viewing data, the server compiles ad viewing rates and effectiveness indicators, and provides rewards to content creators based on the viewing data. These rewards are credited to the content creator's account as electronic money or points.

[0767] Specific examples

[0768] For example, consider the case where a user uploads a travel vlog video. The server analyzes the video frame by frame and identifies scenes such as "introductory scenes to tourist attractions" and "tasting local cuisine." Next, it selects advertising materials for sightseeing tours related to the identified "introduction scenes" and automatically generates a 15-second advertising video. This advertising video is spliced ​​in immediately after the "introduction scenes to tourist attractions" so that it appears natural to the viewer.

[0769] When viewers play the modified videos, the server tracks viewing data, which is then used to provide compensation to content creators, enabling them to generate direct revenue and businesses to more effectively deliver advertising.

[0770] Prompt Sentence Examples

[0771] "Upload a travel vlog video, analyze it, and automatically generate an ad video related to that scene. For example, insert an ad for a sightseeing tour into a scene introducing a tourist spot, and edit it so that it is presented to viewers in a natural flow."

[0772] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0773] Step 1:

[0774] Users upload video files from their devices using a dedicated website or application. At this time, users input the necessary metadata (video title, description, etc.) along with the video file. The input data, which consists of the video file and metadata, is then sent to the server.

[0775] Step 2:

[0776] The server receives the video file and metadata sent by the user and stores them in a database. The input data is the uploaded video file and metadata, and they are stored in the database.

[0777] Step 3:

[0778] The server uses FFmpeg to break down the stored video into frames and extract the image data for each frame. This process converts the input video file into a series of image data. The output data is the image data for each frame.

[0779] Step 4:

[0780] The server uses OpenCV and TensorFlow to analyze the image data of each frame. For example, it performs processing to identify people, objects, backgrounds, etc. The input data is the image data for each frame, and the output data is analyzed object information (for example, "people," "scenery," "buildings," etc.).

[0781] Step 5:

[0782] The server converts the audio data of the video into text using Google Cloud Speech-to-Text. Through this process, the input data is the audio data in the video, and the output data is the converted text data.

[0783] Step 6:

[0784] The server identifies scenes and topics based on the results of analyzing the image data and audio data. For example, it identifies scenes such as "introductions to tourist spots" or "tasting local cuisine." The input data are the results of analyzing the image data and audio data, and the output data is information about the identified scenes and topics.

[0785] Step 7:

[0786] The server selects relevant advertising materials from a database based on the identified scene and topic. This selection takes into account the relevance to the scene or topic and the preferences of the target audience. The input data is information about the scene or topic, and the output data is the selected advertising materials.

[0787] Step 8:

[0788] The server combines the selected advertising materials and automatically generates an advertising video using FFmpeg and Adobe Premiere Pro API. For example, it edits the video so that an advertisement for a sightseeing tour seamlessly follows a scene introducing a tourist spot. The input data is the advertising materials, and the output data is the generated advertising video.

[0789] Step 9:

[0790] The server creates a modified video by seamlessly splicing the generated advertising video into the original video. In this process, the input data is the original video and the generated advertising video, and the output data is the modified video.

[0791] Step 10:

[0792] The server delivers the modified video file to the user's device, and at the same time adds metadata (advertising information, scene information, etc.) to the modified video. The input data is the modified video and metadata, and the output data is the modified video to be delivered.

[0793] Step 11:

[0794] The user starts playing the modified video on the device. The server tracks the user's viewing behavior in real time and records the playback status of the advertisement. The input data is the user's viewing behavior, and the output data is the viewing data.

[0795] Step 12:

[0796] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing data, and provides rewards to content creators based on the viewing data. The input data is the viewing data, and the output data is the compilation results and rewards. Rewards are credited to the content creator's account as electronic money or points.

[0797] (Application example 1)

[0798] 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."

[0799] With conventional video advertising systems, the content of ads often did not match the video scenes or topics, causing viewers to feel uncomfortable. Furthermore, they were unable to analyze viewing data to maximize advertising effectiveness and real-time ad repositioning, which resulted in insufficient advertising effectiveness. Furthermore, they were also unable to provide appropriate rewards based on viewing data.

[0800] 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.

[0801] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for analyzing the content of the video and identifying scenes and topics, means for selecting relevant advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for seamlessly connecting the generated advertising video to the original video, means for delivering the modified video to the user terminal, means for tracking viewing data in real time, and means for rearranging the advertising video based on the viewing data. This makes it possible to insert appropriate advertisements without causing viewers a sense of incongruity and maximize advertising effectiveness based on the viewing data. It is also possible to provide appropriate rewards.

[0802] A "user terminal" is an electronic device that a user uses to shoot or upload videos.

[0803] A "database" is a system that systematically stores and manages data such as video files and advertising materials.

[0804] "Means for analyzing the content of video" refers to a processing device or software that breaks down a video file into frames and identifies scenes and topics using image and voice recognition.

[0805] "Means for identifying scenes and topics" refers to technology that analyzes image data and audio data to identify specific scenes and topics within a video.

[0806] The "means for selecting advertising material" is a technique for selecting relevant advertising content from an advertising database based on a specified scene or topic.

[0807] The "means for automatically generating advertising videos" is a video editing algorithm that naturally connects selected advertising materials to the original video.

[0808] The "means for distributing modified videos" refers to a technology that transmits video files with inserted advertisements to users' terminals.

[0809] "Means for tracking viewing data in real time" refers to technology that records the actions of users when they view modified videos in real time.

[0810] The "means for rearranging advertising videos" is a technology that rearranges advertisements to optimal positions based on viewing data.

[0811] A "reward mechanism" is a system or algorithm for rewarding content creators based on aggregated viewing data.

[0812] This invention is a system that uses AI to automatically generate advertisements appropriate for the situation for videos provided by users, and provides advertisements to viewers in a natural way. This system is composed of a user terminal, a server, and a database.

[0813] First, a video file is uploaded from the user's device to the server. The server then stores this video file in a database. The server then breaks down the stored video into frames and extracts image and audio data. The server then uses an image recognition algorithm to analyze the content of each frame (people, objects, background, etc.), and uses a speech recognition algorithm to convert the audio data into text. This analysis can be performed using Google Cloud's Speech-to-Text API or an open-source image recognition library.

[0814] Next, the server selects relevant advertising materials from an advertising database based on the identified scenes and topics. Selection criteria include relevance to the scene or topic, and the preferences of the target audience. The server combines the selected advertising materials and automatically generates an advertising video using libraries such as MoviePy. The generated advertising video is edited to seamlessly integrate with the flow of the original video.

[0815] The modified video file is delivered to the user's terminal by the server. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video. The user plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements. The viewing data is periodically compiled, and advertising viewing rates and effectiveness indicators are calculated. Based on this viewing data, the server can also reposition the advertising videos.

[0816] Furthermore, the server provides rewards to content creators based on the collected viewing data, which are credited to their accounts as electronic money or points.

[0817] For example, if a user uploads a "travel vlog video," the server analyzes the video frame by frame to identify scenes such as "introductory tourist spots" and "tasting local cuisine." The server then selects advertising materials for sightseeing tours related to the identified "introduction of tourist spots" scene and automatically generates a 15-second advertising video. The generated advertising video is spliced ​​in immediately after the "introduction of tourist spots" scene, creating a natural flow for viewers. When viewers play this modified video, the server tracks viewing data. Content creators are then rewarded based on the viewing data.

[0818] Example prompt for a generative AI model:

[0819] Perform scene analysis on user-provided videos, generate and insert appropriate ads. User video: Travel Vlog Scene 1: Tourist attraction introduction Scene 2: Tasting local cuisine Select ads related to the scene from the ad database, and edit the video to make it look natural.

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

[0821] Step 1:

[0822] A user shoots or uploads a video through a smartphone app. The input is the video file uploaded by the user, and this video file is sent to the server. The server stores the received video file in a database. The output is the stored video file.

[0823] Step 2:

[0824] The server breaks down the stored video file into frames. The input is a video file, and the server converts this video into image data for each frame. The output is image data for each frame. This is done using libraries such as OpenCV.

[0825] Step 3:

[0826] The server uses an image recognition algorithm to analyze the image data for each frame and identify people, objects, backgrounds, etc. The input is the image data for each frame, and the output is tag information that indicates the content of each frame. Here, an open-source image recognition library is used.

[0827] Step 4:

[0828] The server uses a speech recognition algorithm to convert the audio data from the video into text. The input is the audio data from the video file, and the output is text data. This is done using Google Cloud's Speech-to-Text API.

[0829] Step 5:

[0830] The server identifies scenes and topics from image and audio data. The input is tag information and text data for each frame, and the output is the identified scene and topic information. A generative AI model is used to extract features from each scene and identify the topic.

[0831] Step 6:

[0832] The server selects relevant advertising materials based on the identified scenes and topics. The input is scene and topic information, and the output is advertising materials retrieved from an advertising database. Selection criteria take into account relevance to the scene or topic, preferences of the target audience, etc.

[0833] Step 7:

[0834] The server automatically generates an advertising video using the selected advertising materials. The input is the advertising materials and scene information, and the output is the generated advertising video. Video editing is performed using libraries such as MoviePy.

[0835] Step 8:

[0836] The server then seamlessly stitches the generated ad video back into the original video. The input is the original video and the ad video, and the output is the modified video. This process is also performed using libraries such as MoviePy.

[0837] Step 9:

[0838] The server delivers the modified video file to the user's device. The input is the modified video, and the output is the video sent to the user's device. At this time, metadata such as added advertising information and scene information is also delivered.

[0839] Step 10:

[0840] The server tracks the user's viewing behavior in real time when they play the modified video. The input is the user's viewing behavior data, and the output is the tracked viewing data. Real-time tracking is performed using sensors and log data.

[0841] Step 11:

[0842] The server repositions the ads based on the collected viewing data. The input is the tracked viewing data, and the output is the repositioned ad video.

[0843] Step 12:

[0844] The server provides rewards to content creators based on the aggregated viewing data. The input is viewing data and advertising effectiveness indicators, and the output is electronic money or points given as rewards.

[0845] 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.

[0846] System Overview

[0847] This invention is a system that uses AI to automatically generate advertisements appropriate for the situation for videos provided by users, recognizes users' emotions, and utilizes the data. This system is composed of a user terminal, a server, an emotion engine, and a database.

[0848] Specific operation of the system

[0849] Video upload and analysis

[0850] Users upload video files from their devices through a dedicated website or application. The server receives the video files and stores them in a database. The server then breaks down the stored video into frames and extracts the image data for each frame. The server uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.), and then uses speech recognition algorithms to convert the audio data into text. This allows the server to identify the scenes and topics in the video.

[0851] Selecting advertising materials and generating advertising videos

[0852] Based on the identified scenes and topics, the server selects relevant advertising materials from the advertising provider's database. Selection criteria include the relevance to the scene or topic, and the preferences of the target audience. The server then combines the selected advertising materials and automatically generates an advertising video using a video editing algorithm. The generated advertising video is edited to seamlessly integrate with the flow of the original video.

[0853] User emotion recognition and data utilization

[0854] While the video is playing, the server uses an emotion engine to analyze video and audio data obtained from the user's device in real time and collects user emotional data. This emotional data is measured based on the user's facial expressions, tone of voice, eye movements, etc. The server then analyzes the acquired emotional data and selects the most appropriate advertising material based on that data.

[0855] Video streaming and viewing

[0856] The modified video file is then delivered to the user's device by the server. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video. The user then plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements.

[0857] Aggregating viewing and emotion data and providing rewards

[0858] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing and emotional data. As a result, rewards based on the viewing and emotional data are provided to content creators. The server credits these rewards to the content creators' accounts as electronic money or points.

[0859] Specific examples

[0860] For example, consider the case where a user uploads a travel vlog video. The server analyzes this video frame by frame and identifies scenes such as "introductory tourist spots" and "tasting local cuisine." Next, it selects advertising materials for sightseeing tours related to the identified "introduction of tourist spots" scenes, and uses an emotion engine to select tour information that is likely to interest the user. It automatically generates a 15-second advertising video, optimizing it based on the user's emotional data.

[0861] When viewers play the modified videos, the server captures and tracks emotional data through the emotion engine. Content creators are then rewarded based on the viewing and emotional data, enabling direct revenue for content creators and enabling businesses to deliver effective ad delivery and emotional ad engagement.

[0862] In this way, the present invention is a system that simultaneously reduces corporate advertising costs, improves advertising effectiveness, reduces viewer stress, and provides rewards to content creators.

[0863] The processing flow will be explained below.

[0864] Step 1:

[0865] Users upload video files from their devices via a dedicated website or application, and the server receives the video files and stores them in a database.

[0866] Step 2:

[0867] The server breaks down the stored video into frames and extracts the image data for each frame.The server then uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.).

[0868] Step 3:

[0869] The server extracts audio data from the video, converts it into text using a speech recognition algorithm, and then analyzes the text data to identify scenes and topics.

[0870] Step 4:

[0871] The server selects relevant advertising materials from the database of advertising providers based on the identified scenes and topics, taking into account the relevance to the scenes and topics and the preferences of the target audience as selection criteria.

[0872] Step 5:

[0873] The server uses the selected advertising materials and a video editing algorithm to automatically generate a 15-second advertising video.The server then edits the generated advertising video so that it fits naturally into the original video without disrupting the flow of the original video.

[0874] Step 6:

[0875] While the video is playing, the server uses an emotion engine to analyze video and audio data obtained from the user's device in real time and collects emotional data from the user, which is measured based on facial expressions, tone of voice, eye movements, etc.

[0876] Step 7:

[0877] The server analyzes the acquired emotional data and selects the most suitable advertising material based on that data. If the emotional data indicates joy, it performs optimization processing such as selecting advertising material with more positive content.

[0878] Step 8:

[0879] The server adds metadata (advertising information, scene information, etc.) to the modified video file and provides a link and download option for distribution to the user's device.

[0880] Step 9:

[0881] The user plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the ad playback status (start, stop, skip, etc.).

[0882] Step 10:

[0883] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing and emotion data.The server then processes the content creators to receive rewards based on the viewing and emotion data.The rewards are credited to the content creators' accounts as electronic money or points.

[0884] Step 11:

[0885] The server generates detailed reports for advertising companies on the effectiveness of their ads based on viewing data and sentiment data, allowing them to analyze the cost-effectiveness of their ads and provide feedback for their next advertising strategy and optimization.

[0886] Example 2

[0887] 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."

[0888] Conventional advertising systems only select ads based on video scenes and topics, making it difficult to optimize ads according to user emotions. Furthermore, there were no systems that could collect and analyze user emotions in real time and reselect advertising materials based on that information. This limited the effectiveness of advertising and made it difficult to improve user engagement.

[0889] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0890] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for analyzing the content of the video and identifying scenes and topics, means for selecting related advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for naturally splicing the generated advertising video into the original video, means for delivering the modified video to the user terminal, means for collecting and analyzing user emotion data in real time, and means for reselecting advertising materials based on the user emotion data, thereby enabling advertising delivery optimized to the user's emotions.

[0891] A "user terminal" is a device used by a user to upload and play videos, and includes smartphones, tablets, personal computers, etc.

[0892] A "server" is a computer system that is responsible for receiving, storing, analyzing, generating and distributing advertising materials, and other processes for video files.

[0893] A "database" is a system for systematically storing and managing data such as video files, advertising materials, and analysis results.

[0894] A "video file" is a file containing video and audio data uploaded by a user.

[0895] "Analysis" is the process of breaking down video content frame by frame, identifying its content using image and voice recognition technology, and identifying scenes and topics.

[0896] A "scene" refers to a specific situation or situation within a video, and is a part of a video that consists of elements such as people, objects, and background.

[0897] "Topics" refer to the themes or topics covered in the video, such as introductions to tourist spots or product reviews.

[0898] "Advertising materials" refers to elements such as images, videos, and audio used for advertising distribution, and are promotional content provided by companies.

[0899] "Automatic generation" refers to the process by which a system programmatically generates advertising videos without manual intervention.

[0900] A "modified video" is a video file that has been edited by adding advertising material to the original video.

[0901] "Emotion data" refers to data relating to the user's emotional state extracted from facial expressions, tone of voice, eye movements, and the like.

[0902] "Collection" refers to the process of acquiring emotion data and viewing data from user terminals and storing them on a server.

[0903] "Analysis" is the process of identifying the user's emotional state based on collected emotional data and selecting advertising materials accordingly.

[0904] "Viewing data" refers to information about the user's actions when watching a video, including viewing time, number of views, clicks, and the like.

[0905] "Reward" refers to compensation such as electronic money or points provided to content creators based on viewing data and emotional data.

[0906] The present invention is a system that automatically generates advertisements appropriate for the situation for videos provided by users, recognizes the user's emotions, and utilizes the data. This system is composed of a user terminal, a server, an emotion engine, and a database.

[0907] First, a user uploads a video file through a dedicated website or application. The device used can be a smartphone, tablet, or personal computer. The video file is then received by a server and stored in a database. Specifically, the database may use SQL Server or MySQL.

[0908] The server then breaks down the stored video into frames and extracts the image data for each frame. It uses image recognition algorithms such as OpenCV and speech recognition algorithms such as the Google Speech-to-Text API. The server then uses these algorithms to identify scenes and topics in the video.

[0909] Based on the identified scenes and topics, the server selects relevant advertising materials from the advertising provider's database. Selection criteria include relevance to the scene or topic, and the preferences of the target audience. The server then combines the selected advertising materials and automatically generates an advertising video using a video editing algorithm (e.g., FFmpeg).

[0910] After generating the ad video, the server uses an emotion engine to collect video and audio data from the user's device in real time to obtain the user's emotional data. Microsoft Azure Cognitive Services is used to analyze the emotional data. By analyzing the collected emotional data, the server can understand the user's emotional state while watching and can reselect advertising materials based on that data.

[0911] The modified video file is then delivered to the user's device by the server. This delivery utilizes streaming technology, such as RTMP (Real-Time Message Protocol). Metadata is added to the modified video, and user viewing status is tracked. The user then plays the modified video on their device and begins watching.

[0912] Finally, the server compiles ad viewing rates and effectiveness indicators based on the collected viewing and emotion data. Based on the compilation results, rewards are provided to content creators using the viewing and emotion data. Rewards are credited to the content creators' accounts as electronic money or points.

[0913] As a concrete example, consider the case where a user uploads a travel vlog video. The server analyzes the video and identifies scenes such as "introducing tourist spots" and "tasting local cuisine." Next, it selects advertising materials related to these scenes and seamlessly stitches the automatically generated advertising videos into the original video. When a user plays this modified video, the server obtains emotional data in real time and delivers optimal advertising, thereby improving user engagement.

[0914] Example prompt: "Describe how an AI system can automatically generate relevant ads for travel vlogs that showcase tourist attractions when uploaded."

[0915] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0916] Step 1:

[0917] A user uploads a video.

[0918] Input: A video file selected by the user.

[0919] Output: The video file that will be sent to the server.

[0920] How it works: The user selects a video file through a dedicated website or application and clicks the upload button. The video file is then sent to the server via the Internet.

[0921] Step 2:

[0922] The server receives the video and stores it in a database.

[0923] Input: The video file submitted by the user.

[0924] Output: Video files stored in the database.

[0925] Specific operation: The server receives the video file and saves it in a database (e.g., MySQL or SQL Server). When saving, the video file's metadata (e.g., file name, upload date, etc.) is also registered.

[0926] Step 3:

[0927] The server analyzes the video.

[0928] Input: Video files stored in the database.

[0929] Output: Image data decomposed into frames, and audio data converted to text.

[0930] What it does: The server breaks down the video file into frames, extracts the image data for each frame using OpenCV, and converts the audio data to text using the Google Speech-to-Text API.

[0931] Step 4:

[0932] The server identifies the scene and topic.

[0933] Input: Image data decomposed frame by frame and audio data converted to text.

[0934] Output: Data about identified scenes and topics.

[0935] How it works: The server uses image recognition and text analysis algorithms to identify scenes (e.g., introductions to tourist attractions, tastings of local cuisine) and topics within the video.

[0936] Step 5:

[0937] The server selects the advertising material.

[0938] Input: Data about the identified scene and topic.

[0939] Output: Selected advertising material data.

[0940] Specific operation: Based on the analysis results, the server issues a query to the ad provider's database to obtain relevant ad material data. Selection criteria take into account the relevance to the scene or topic, the preferences of the target audience, etc.

[0941] Step 6:

[0942] The server generates the advertisement video.

[0943] Input: Selected advertising material data.

[0944] Output: Auto-generated ad video.

[0945] Specific operation: The server combines the advertising materials and automatically generates an advertising video using a video editing algorithm (e.g., FFmpeg). The generated advertising video is edited to seamlessly connect with the original video.

[0946] Step 7:

[0947] The server collects and analyzes the emotion data.

[0948] Input: Real-time video and audio data sent from the user terminal.

[0949] Output: Parsed emotion data.

[0950] How it works: The server uses an emotion engine (such as Microsoft Azure Cognitive Services) to collect and analyze the user's video and audio data in real time. The analysis results include the user's facial expressions, tone of voice, and eye movements.

[0951] Step 8:

[0952] The server delivers the video to the user's device.

[0953] Input: The modified video file.

[0954] Output: A streaming URL that can be played on the user's device.

[0955] Specific operation: The server uploads the modified video file to the streaming service, generates a streaming URL, and notifies the user of this URL.

[0956] Step 9:

[0957] The server collects viewing data and provides rewards.

[0958] Input: User viewing and emotion data.

[0959] Output: Aggregated advertising effectiveness metrics and rewards for content creators.

[0960] Specific operation: The server monitors video playback status and emotional data in real time, aggregates viewing data, and awards electronic money or points to content creators through a reward system based on the aggregated results.

[0961] (Application example 2)

[0962] 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."

[0963] Conventional advertising systems lack the ability to automatically select and generate effective advertising materials for video content provided by users, making it difficult to deliver ads that take into account user emotions and the content of scenes. Furthermore, technology for optimizing ads based on viewer emotions is immature, and compensation for content creators is limited to viewing data. The purpose of this invention is to solve these issues and maximize advertising effectiveness while providing excellent convenience for viewers and content creators.

[0964] 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.

[0965] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for analyzing the content of the video and identifying scenes and topics, means for selecting related advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for naturally splicing the generated advertising video into the original video, means for delivering the modified video to the user terminal, and means for analyzing user emotions in real time and optimizing the advertising materials using the data. This makes it possible to analyze user emotions and video scenes in real time and automatically generate and deliver optimized advertising videos.

[0966] A "user terminal" is a device used by a user, such as a computer, smartphone, or tablet.

[0967] A "video file" is a digital file containing video and audio provided by a user.

[0968] A "database" is a system for storing and managing video files and related information.

[0969] The "server" is a centralized management system for managing the database, analyzing videos, and generating advertisements.

[0970] A "scene" refers to a specific situation or situation within a video.

[0971] "Topics" are themes or topics covered in the video.

[0972] "Advertising materials" are content such as video, audio, and text that are the source of the advertisement to be distributed.

[0973] An "advertising video" is an advertising video inserted into a video provided by a user.

[0974] "Emotion" refers to the emotional response or state that a user shows when watching a video.

[0975] "Advertising optimization" is the act of selecting and editing optimal advertising materials based on user emotions and video scenes.

[0976] "Viewing data" is data that records the actions and reactions of users when they watch videos.

[0977] "Reward" refers to monetary or points provided to content creators based on viewing data and emotional data.

[0978] To implement this invention, several major pieces of hardware and software are required, such as a server, a user terminal, a database, and an emotion engine.

[0979] First, a user shoots a video file using their own device and uploads it to the server via a dedicated application or website. User devices can be PCs, smartphones, tablets, etc. The server receives the uploaded video file and stores it in a database. The database centrally manages video files and their metadata.

[0980] The server then breaks down the saved video file into frames and analyzes the image and audio data of each frame. This process utilizes image and audio recognition algorithms. Specifically, OpenCV is used to analyze the image frames, and DeepFace and other algorithms are used to recognize facial expressions. Natural language processing libraries (such as the Hugging Face transformer) are used for audio recognition.

[0981] The server analyzes the content of the video to identify its scenes and topics, then selects advertising materials appropriate for the scenes from a related database, taking into account the content of the scenes and the attributes of the target audience.

[0982] Once the ad material selection is complete, the server automatically generates an ad video using the selected ad material. At this time, a video editing library (e.g., MoviePy) is used to seamlessly stitch the ad video into the original video. The generated ad video is designed to maintain the flow of the original video.

[0983] The modified video is then sent back to the user's device from the server. The user watches the modified video, and during this process the server analyzes the user's emotions in real time. DeepFace and other emotion analysis engines are used as emotion engines. For example, the server collects user emotional data based on eye movements, facial expressions, and tone of voice.

[0984] Furthermore, the system tracks the viewing status of the modified videos and aggregates the viewing data. This viewing data and emotional data are used to provide compensation to content creators. The compensation is credited to the content creator's account in the form of electronic money or points.

[0985] For example, suppose a user uploads a travel vlog video. In this case, the server analyzes the video and identifies scenes such as "introducing tourist spots" and "tasting local cuisine." It then selects advertising materials for sightseeing tours that are suitable for the "introducing tourist spots" scene and automatically generates a 15-second advertising video based on them. The generated advertising video is then seamlessly integrated into the original video and optimized based on the user's emotional data.

[0986] An example of a prompt sentence is as follows:

[0987] Generate a Python program that performs frame-based emotion recognition and inserts appropriate ads.

[0988] By using such prompts, it is possible to have a generative AI model automatically generate a program to perform a specific task.

[0989] This makes it possible to analyze user emotions and video scenes in real time, and automatically generate and deliver optimized advertising videos.

[0990] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0991] Step 1:

[0992] Upload video files from the user's device

[0993] Users use their own devices to upload video files to the server through a dedicated application or website. At this time, the video files are stored in a database. The input is the video file recorded by the user, and the output is the video data stored in the database.

[0994] Step 2:

[0995] Video decomposition and frame analysis

[0996] The server breaks down the stored video file into frames. It uses OpenCV to analyze the image data of each frame and recognizes faces, objects, etc. The input is the video file stored in the database, and the output is the image data of each frame. Specifically, it uses an image recognition algorithm to analyze people, backgrounds, etc. in the video.

[0997] Step 3:

[0998] Audio analysis

[0999] The server uses a speech recognition algorithm to analyze the audio data in the video file, convert it into text, and identify the scenes and topics in the video. The input is the audio data for each frame, and the output is text data.

[1000] Step 4:

[1001] Identifying scenes and topics

[1002] The server integrates image data and audio data to identify scenes and topics within the video. Natural language processing algorithms and emotion recognition models are used here. The input is image data and audio text data for each frame, and the output is information on the identified scenes and topics.

[1003] Step 5:

[1004] Selection of advertising materials

[1005] The server selects the most suitable advertising material from a database based on the identified scene and topic, taking into account factors such as the relevance to the scene and the preferences of the target audience. The input is information about the scene and topic, and the output is the selected advertising material.

[1006] Step 6:

[1007] Ad video generation

[1008] The server automatically generates an ad video using the selected ad material. Using a video editing library such as MoviePy, the ad video is seamlessly spliced ​​into the original video. The input is the original video frames and ad material, and the output is the generated ad video.

[1009] Step 7:

[1010] Distribution of altered videos

[1011] The server integrates the generated advertising video with the original video and delivers the modified video file to the user terminal. The input is the generated advertising video and the original video, and the output is the modified video file.

[1012] Step 8:

[1013] User sentiment analysis

[1014] When a user watches the modified video, the server uses an emotion engine to analyze the user's reactions (facial expressions, gaze, tone of voice, etc.). The input is the video and audio data the user is watching, and the output is the user's emotional data.

[1015] Step 9:

[1016] Tracking and aggregating viewing data

[1017] The server tracks the viewing status of the modified video and aggregates the viewing data. The input is user viewing behavior data, and the output is aggregated viewing data.

[1018] Step 10:

[1019] Providing rewards

[1020] The server provides rewards to content creators based on the viewing and emotion data. The rewards are added to their accounts as electronic money or points. The input is the aggregated viewing and emotion data, and the output is the rewards to the content creators.

[1021] 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.

[1022] 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.

[1023] 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.

[1024] [Fourth embodiment]

[1025] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1026] 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.

[1027] 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).

[1028] 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.

[1029] 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.

[1030] 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).

[1031] 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. 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.

[1032] 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.

[1033] 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.

[1034] 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.

[1035] 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.

[1036] 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.

[1037] 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."

[1038] System Overview

[1039] This invention is a system that uses AI to automatically generate advertisements appropriate for the situation for videos provided by users, and provides advertisements to viewers in a natural way. This system is composed of a user terminal, a server, and a database.

[1040] Specific operation of the system

[1041] Video upload and analysis

[1042] Users upload video files from their devices through a dedicated website or application. The server receives the video files and stores them in a database. The server then breaks down the stored video into frames and extracts the image data for each frame. The server uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.), and then uses speech recognition algorithms to convert the audio data into text. This allows the server to identify the scenes and topics in the video.

[1043] Selecting advertising materials and generating advertising videos

[1044] Based on the identified scenes and topics, the server selects relevant advertising materials from the advertising provider's database. Selection criteria include the relevance to the scene or topic, and the preferences of the target audience. The server then combines the selected advertising materials and automatically generates an advertising video using a video editing algorithm. The generated advertising video is edited to seamlessly integrate with the flow of the original video.

[1045] Video streaming and viewing

[1046] The modified video file is then delivered to the user's device by the server. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video. The user then plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements.

[1047] Aggregating viewing data and providing rewards

[1048] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing data. This allows content creators to receive rewards based on the viewing data. The server then credits these rewards to the content creators' accounts as electronic money or points.

[1049] Specific examples

[1050] For example, consider the case where a user uploads a travel vlog video. The server analyzes the video frame by frame and identifies scenes such as "introductory tourist spots" and "tasting local cuisine." Next, it selects advertising materials for sightseeing tours related to the identified "introduction of tourist spots" scene and automatically generates a 15-second advertising video. The generated advertising video is spliced ​​in immediately after the "introduction of tourist spots" scene, so that it is displayed in a natural flow for the viewer.

[1051] When viewers play the modified videos, the server tracks viewing data, which is then used to provide compensation to content creators, enabling them to generate direct revenue and businesses to more effectively deliver advertising.

[1052] In this way, the present invention is a system that simultaneously reduces corporate advertising costs, improves advertising effectiveness, reduces viewer stress, and provides rewards to content creators.

[1053] The processing flow will be explained below.

[1054] Step 1:

[1055] Users upload video files from their devices via a dedicated website or application, and the server receives the video files and stores them in a database.

[1056] Step 2:

[1057] The server breaks down the stored video into frames and extracts the image data for each frame.The server then uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.).

[1058] Step 3:

[1059] The server extracts audio data from the video, converts it into text using a speech recognition algorithm, and then analyzes the text data to identify topics and scenes.

[1060] Step 4:

[1061] The server selects relevant advertising materials from the database of advertising providers based on the identified scenes and topics, taking into account the relevance to the scenes and topics and the preferences of the target audience as selection criteria.

[1062] Step 5:

[1063] The server uses the selected advertising materials and a video editing algorithm to automatically generate a 15-second advertising video.The server then edits the generated advertising video so that it fits naturally into the original video without disrupting the flow of the original video.

[1064] Step 6:

[1065] The server adds metadata (advertising information, scene information, etc.) to the modified video file and provides a link and download option for distribution to the user's device.

[1066] Step 7:

[1067] The user plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the ad playback status (start, stop, skip, etc.).

[1068] Step 8:

[1069] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing data.The server then processes the content creators to provide rewards based on the viewing data.The rewards are credited to the content creators' accounts as electronic money or points.

[1070] Step 9:

[1071] The server generates detailed reports on ad viewing data and effectiveness for ad providers, allowing them to analyze the cost-effectiveness of advertising and provide feedback for future advertising strategies and optimization.

[1072] Example 1

[1073] 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."

[1074] Conventional ad distribution systems have struggled to insert ads into video content at appropriate and natural timing. Furthermore, many ads are intrusive to viewers, and compensation mechanisms for content creators are often unclear. This leads to lower viewer satisfaction and insufficient advertising effectiveness. Furthermore, there is a need for accurate tracking of viewing data and the provision of appropriate compensation to content creators.

[1075] 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.

[1076] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for breaking down the video into frames and extracting image data for each frame, means for analyzing the extracted image data to identify content such as people, objects, and backgrounds, means for converting audio data into text to identify scenes and topics, means for selecting relevant advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for naturally splicing the generated advertising video into the original video, and means for delivering the modified video to the user terminal. This makes it possible to insert advertisements into video content in a natural way and increase viewer satisfaction. It also makes it possible to accurately track viewing data and provide compensation to content creators based on the viewing data.

[1077] "User terminal" refers to a device used by a user, such as a computer, smartphone, or tablet.

[1078] A "video file" refers to a digital file format in which video and audio are recorded.

[1079] A "database" refers to a system for efficiently managing and storing data.

[1080] A "frame" refers to an individual still image that makes up a video.

[1081] "Image data" refers to digitized data of the visual information contained in each frame.

[1082] "Image recognition algorithms" refer to techniques that allow computers to understand objects and scenes in images.

[1083] A "voice recognition algorithm" refers to the technology that analyzes voice data and converts it into text.

[1084] A "scene" is a unit that represents a specific situation or situation within a video.

[1085] "Topics" refer to the themes or topics touched upon in the video.

[1086] "Advertising Materials" means information or content used for advertising.

[1087] "Video editing algorithm" refers to technology that automatically performs editing such as joining videos, trimming, and adding effects.

[1088] "Altered videos" refer to videos that have been edited by inserting advertisements into the original video.

[1089] "Metadata" refers to data about data, i.e., additional information about the original data.

[1090] "Tracking" refers to a system tracking user actions and data.

[1091] "Viewing data" refers to information about a user's viewing of a video (such as playback time, whether or not the video was skipped, and whether or not the ad was viewed).

[1092] "Reward" refers to the compensation provided by the system to content creators.

[1093] "Content creator" refers to an individual or organization that produces digital content such as videos.

[1094] This invention is a system that automatically generates advertisements appropriate for the situation for videos provided by users and provides advertisements to viewers in a natural way. This system is composed of a user terminal, a server, and a database.

[1095] System Overview

[1096] The server receives video files from user devices and stores them in a database. The server then breaks down the video into frames and extracts the image data for each frame. Software used includes OpenCV and TensorFlow. This allows the server to analyze the image data for each frame and identify content such as people, objects, and background. Google Cloud Speech-to-Text can also be used to convert audio data contained in the video into text and identify scenes and topics.

[1097] Based on the identified scenes and topics, the server selects relevant advertising materials from a database. The software used for this purpose includes, for example, a link to the advertising provider's database and custom scripts to implement the selection algorithm. The server then combines the selected advertising materials and automatically generates an advertising video using FFmpeg or Adobe Premiere Pro APIs.

[1098] The generated advertising video is then seamlessly integrated into the original video and delivered to the user's device. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video file.

[1099] The user plays the modified video on their device and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements. Based on the collected viewing data, the server compiles ad viewing rates and effectiveness indicators, and provides rewards to content creators based on the viewing data. These rewards are credited to the content creator's account as electronic money or points.

[1100] Specific examples

[1101] For example, consider the case where a user uploads a travel vlog video. The server analyzes the video frame by frame and identifies scenes such as "introductory scenes to tourist attractions" and "tasting local cuisine." Next, it selects advertising materials for sightseeing tours related to the identified "introduction scenes" and automatically generates a 15-second advertising video. This advertising video is spliced ​​in immediately after the "introduction scenes to tourist attractions" so that it appears natural to the viewer.

[1102] When viewers play the modified videos, the server tracks viewing data, which is then used to provide compensation to content creators, enabling them to generate direct revenue and businesses to more effectively deliver advertising.

[1103] Prompt Sentence Examples

[1104] "Upload a travel vlog video, analyze it, and automatically generate an ad video related to that scene. For example, insert an ad for a sightseeing tour into a scene introducing a tourist spot, and edit it so that it is presented to viewers in a natural flow."

[1105] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1106] Step 1:

[1107] Users upload video files from their devices using a dedicated website or application. At this time, users input the necessary metadata (video title, description, etc.) along with the video file. The input data, which consists of the video file and metadata, is then sent to the server.

[1108] Step 2:

[1109] The server receives the video file and metadata sent by the user and stores them in a database. The input data is the uploaded video file and metadata, and they are stored in the database.

[1110] Step 3:

[1111] The server uses FFmpeg to break down the stored video into frames and extract the image data for each frame. This process converts the input video file into a series of image data. The output data is the image data for each frame.

[1112] Step 4:

[1113] The server uses OpenCV and TensorFlow to analyze the image data of each frame. For example, it performs processing to identify people, objects, backgrounds, etc. The input data is the image data for each frame, and the output data is analyzed object information (for example, "people," "scenery," "buildings," etc.).

[1114] Step 5:

[1115] The server converts the audio data of the video into text using Google Cloud Speech-to-Text. Through this process, the input data is the audio data in the video, and the output data is the converted text data.

[1116] Step 6:

[1117] The server identifies scenes and topics based on the results of analyzing the image data and audio data. For example, it identifies scenes such as "introductions to tourist spots" or "tasting local cuisine." The input data are the results of analyzing the image data and audio data, and the output data is information about the identified scenes and topics.

[1118] Step 7:

[1119] The server selects relevant advertising materials from a database based on the identified scene and topic. This selection takes into account the relevance to the scene or topic and the preferences of the target audience. The input data is information about the scene or topic, and the output data is the selected advertising materials.

[1120] Step 8:

[1121] The server combines the selected advertising materials and automatically generates an advertising video using FFmpeg and Adobe Premiere Pro API. For example, it edits the video so that an advertisement for a sightseeing tour seamlessly follows a scene introducing a tourist spot. The input data is the advertising materials, and the output data is the generated advertising video.

[1122] Step 9:

[1123] The server creates a modified video by seamlessly splicing the generated advertising video into the original video. In this process, the input data is the original video and the generated advertising video, and the output data is the modified video.

[1124] Step 10:

[1125] The server delivers the modified video file to the user's device, and at the same time adds metadata (advertising information, scene information, etc.) to the modified video. The input data is the modified video and metadata, and the output data is the modified video to be delivered.

[1126] Step 11:

[1127] The user starts playing the modified video on the device. The server tracks the user's viewing behavior in real time and records the playback status of the advertisement. The input data is the user's viewing behavior, and the output data is the viewing data.

[1128] Step 12:

[1129] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing data, and provides rewards to content creators based on the viewing data. The input data is the viewing data, and the output data is the compilation results and rewards. Rewards are credited to the content creator's account as electronic money or points.

[1130] (Application example 1)

[1131] 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."

[1132] With conventional video advertising systems, the content of ads often did not match the video scenes or topics, causing viewers to feel uncomfortable. Furthermore, they were unable to analyze viewing data to maximize advertising effectiveness and real-time ad repositioning, which resulted in insufficient advertising effectiveness. Furthermore, they were also unable to provide appropriate rewards based on viewing data.

[1133] 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.

[1134] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for analyzing the content of the video and identifying scenes and topics, means for selecting relevant advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for seamlessly connecting the generated advertising video to the original video, means for delivering the modified video to the user terminal, means for tracking viewing data in real time, and means for rearranging the advertising video based on the viewing data. This makes it possible to insert appropriate advertisements without causing viewers a sense of incongruity and maximize advertising effectiveness based on the viewing data. It is also possible to provide appropriate rewards.

[1135] A "user terminal" is an electronic device that a user uses to shoot or upload videos.

[1136] A "database" is a system that systematically stores and manages data such as video files and advertising materials.

[1137] "Means for analyzing the content of video" refers to a processing device or software that breaks down a video file into frames and identifies scenes and topics using image and voice recognition.

[1138] "Means for identifying scenes and topics" refers to technology that analyzes image data and audio data to identify specific scenes and topics within a video.

[1139] The "means for selecting advertising material" is a technique for selecting relevant advertising content from an advertising database based on a specified scene or topic.

[1140] The "means for automatically generating advertising videos" is a video editing algorithm that naturally connects selected advertising materials to the original video.

[1141] The "means for distributing modified videos" refers to a technology that transmits video files with inserted advertisements to users' terminals.

[1142] "Means for tracking viewing data in real time" refers to technology that records the actions of users when they view modified videos in real time.

[1143] The "means for rearranging advertising videos" is a technology that rearranges advertisements to optimal positions based on viewing data.

[1144] A "reward mechanism" is a system or algorithm for rewarding content creators based on aggregated viewing data.

[1145] This invention is a system that uses AI to automatically generate advertisements appropriate for the situation for videos provided by users, and provides advertisements to viewers in a natural way. This system is composed of a user terminal, a server, and a database.

[1146] First, a video file is uploaded from the user's device to the server. The server then stores this video file in a database. The server then breaks down the stored video into frames and extracts image and audio data. The server then uses an image recognition algorithm to analyze the content of each frame (people, objects, background, etc.), and uses a speech recognition algorithm to convert the audio data into text. This analysis can be performed using Google Cloud's Speech-to-Text API or an open-source image recognition library.

[1147] Next, the server selects relevant advertising materials from an advertising database based on the identified scenes and topics. Selection criteria include relevance to the scene or topic, and the preferences of the target audience. The server combines the selected advertising materials and automatically generates an advertising video using libraries such as MoviePy. The generated advertising video is edited to seamlessly integrate with the flow of the original video.

[1148] The modified video file is delivered to the user's terminal by the server. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video. The user plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements. The viewing data is periodically compiled, and advertising viewing rates and effectiveness indicators are calculated. Based on this viewing data, the server can also reposition the advertising videos.

[1149] Furthermore, the server provides rewards to content creators based on the collected viewing data, which are credited to their accounts as electronic money or points.

[1150] For example, if a user uploads a "travel vlog video," the server analyzes the video frame by frame to identify scenes such as "introductory tourist spots" and "tasting local cuisine." The server then selects advertising materials for sightseeing tours related to the identified "introduction of tourist spots" scene and automatically generates a 15-second advertising video. The generated advertising video is spliced ​​in immediately after the "introduction of tourist spots" scene, creating a natural flow for viewers. When viewers play this modified video, the server tracks viewing data. Content creators are then rewarded based on the viewing data.

[1151] Example prompt for a generative AI model:

[1152] Perform scene analysis on user-provided videos, generate and insert appropriate ads. User video: Travel Vlog Scene 1: Tourist attraction introduction Scene 2: Tasting local cuisine Select ads related to the scene from the ad database, and edit the video to make it look natural.

[1153] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1154] Step 1:

[1155] A user shoots or uploads a video through a smartphone app. The input is the video file uploaded by the user, and this video file is sent to the server. The server stores the received video file in a database. The output is the stored video file.

[1156] Step 2:

[1157] The server breaks down the stored video file into frames. The input is a video file, and the server converts this video into image data for each frame. The output is image data for each frame. This is done using libraries such as OpenCV.

[1158] Step 3:

[1159] The server uses an image recognition algorithm to analyze the image data for each frame and identify people, objects, backgrounds, etc. The input is the image data for each frame, and the output is tag information that indicates the content of each frame. Here, an open-source image recognition library is used.

[1160] Step 4:

[1161] The server uses a speech recognition algorithm to convert the audio data from the video into text. The input is the audio data from the video file, and the output is text data. This is done using Google Cloud's Speech-to-Text API.

[1162] Step 5:

[1163] The server identifies scenes and topics from image and audio data. The input is tag information and text data for each frame, and the output is the identified scene and topic information. A generative AI model is used to extract features from each scene and identify the topic.

[1164] Step 6:

[1165] The server selects relevant advertising materials based on the identified scenes and topics. The input is scene and topic information, and the output is advertising materials retrieved from an advertising database. Selection criteria take into account relevance to the scene or topic, preferences of the target audience, etc.

[1166] Step 7:

[1167] The server automatically generates an advertising video using the selected advertising materials. The input is the advertising materials and scene information, and the output is the generated advertising video. Video editing is performed using libraries such as MoviePy.

[1168] Step 8:

[1169] The server then seamlessly stitches the generated ad video back into the original video. The input is the original video and the ad video, and the output is the modified video. This process is also performed using libraries such as MoviePy.

[1170] Step 9:

[1171] The server delivers the modified video file to the user's device. The input is the modified video, and the output is the video sent to the user's device. At this time, metadata such as added advertising information and scene information is also delivered.

[1172] Step 10:

[1173] The server tracks the user's viewing behavior in real time when they play the modified video. The input is the user's viewing behavior data, and the output is the tracked viewing data. Real-time tracking is performed using sensors and log data.

[1174] Step 11:

[1175] The server repositions the ads based on the collected viewing data. The input is the tracked viewing data, and the output is the repositioned ad video.

[1176] Step 12:

[1177] The server provides rewards to content creators based on the aggregated viewing data. The input is viewing data and advertising effectiveness indicators, and the output is electronic money or points given as rewards.

[1178] 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.

[1179] System Overview

[1180] This invention is a system that uses AI to automatically generate advertisements appropriate for the situation for videos provided by users, recognizes users' emotions, and utilizes the data. This system is composed of a user terminal, a server, an emotion engine, and a database.

[1181] Specific operation of the system

[1182] Video upload and analysis

[1183] Users upload video files from their devices through a dedicated website or application. The server receives the video files and stores them in a database. The server then breaks down the stored video into frames and extracts the image data for each frame. The server uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.), and then uses speech recognition algorithms to convert the audio data into text. This allows the server to identify the scenes and topics in the video.

[1184] Selecting advertising materials and generating advertising videos

[1185] Based on the identified scenes and topics, the server selects relevant advertising materials from the advertising provider's database. Selection criteria include the relevance to the scene or topic, and the preferences of the target audience. The server then combines the selected advertising materials and automatically generates an advertising video using a video editing algorithm. The generated advertising video is edited to seamlessly integrate with the flow of the original video.

[1186] User emotion recognition and data utilization

[1187] While the video is playing, the server uses an emotion engine to analyze video and audio data obtained from the user's device in real time and collects user emotional data. This emotional data is measured based on the user's facial expressions, tone of voice, eye movements, etc. The server then analyzes the acquired emotional data and selects the most appropriate advertising material based on that data.

[1188] Video streaming and viewing

[1189] The modified video file is then delivered to the user's device by the server. At this time, the server adds metadata (advertising information, scene information, etc.) to the modified video. The user then plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the playback status of the advertisements.

[1190] Aggregating viewing and emotion data and providing rewards

[1191] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing and emotional data. As a result, rewards based on the viewing and emotional data are provided to content creators. The server credits these rewards to the content creators' accounts as electronic money or points.

[1192] Specific examples

[1193] For example, consider the case where a user uploads a travel vlog video. The server analyzes this video frame by frame and identifies scenes such as "introductory tourist spots" and "tasting local cuisine." Next, it selects advertising materials for sightseeing tours related to the identified "introduction of tourist spots" scenes, and uses an emotion engine to select tour information that is likely to interest the user. It automatically generates a 15-second advertising video, optimizing it based on the user's emotional data.

[1194] When viewers play the modified videos, the server captures and tracks emotional data through the emotion engine. Content creators are then rewarded based on the viewing and emotional data, enabling direct revenue for content creators and enabling businesses to deliver effective ad delivery and emotional ad engagement.

[1195] In this way, the present invention is a system that simultaneously reduces corporate advertising costs, improves advertising effectiveness, reduces viewer stress, and provides rewards to content creators.

[1196] The processing flow will be explained below.

[1197] Step 1:

[1198] Users upload video files from their devices via a dedicated website or application, and the server receives the video files and stores them in a database.

[1199] Step 2:

[1200] The server breaks down the stored video into frames and extracts the image data for each frame.The server then uses image recognition algorithms to analyze the content of each frame (people, objects, background, etc.).

[1201] Step 3:

[1202] The server extracts audio data from the video, converts it into text using a speech recognition algorithm, and then analyzes the text data to identify scenes and topics.

[1203] Step 4:

[1204] The server selects relevant advertising materials from the database of advertising providers based on the identified scenes and topics, taking into account the relevance to the scenes and topics and the preferences of the target audience as selection criteria.

[1205] Step 5:

[1206] The server uses the selected advertising materials and a video editing algorithm to automatically generate a 15-second advertising video.The server then edits the generated advertising video so that it fits naturally into the original video without disrupting the flow of the original video.

[1207] Step 6:

[1208] While the video is playing, the server uses an emotion engine to analyze video and audio data obtained from the user's device in real time and collects emotional data from the user, which is measured based on facial expressions, tone of voice, eye movements, etc.

[1209] Step 7:

[1210] The server analyzes the acquired emotional data and selects the most suitable advertising material based on that data. If the emotional data indicates joy, it performs optimization processing such as selecting advertising material with more positive content.

[1211] Step 8:

[1212] The server adds metadata (advertising information, scene information, etc.) to the modified video file and provides a link and download option for distribution to the user's device.

[1213] Step 9:

[1214] The user plays the modified video and begins watching. The server tracks the user's viewing behavior in real time and records the ad playback status (start, stop, skip, etc.).

[1215] Step 10:

[1216] The server compiles ad viewing rates and effectiveness indicators based on the collected viewing and emotion data.The server then processes the content creators to receive rewards based on the viewing and emotion data.The rewards are credited to the content creators' accounts as electronic money or points.

[1217] Step 11:

[1218] The server generates detailed reports for advertising companies on the effectiveness of their ads based on viewing data and sentiment data, allowing them to analyze the cost-effectiveness of their ads and provide feedback for their next advertising strategy and optimization.

[1219] Example 2

[1220] 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."

[1221] Conventional advertising systems only select ads based on video scenes and topics, making it difficult to optimize ads according to user emotions. Furthermore, there were no systems that could collect and analyze user emotions in real time and reselect advertising materials based on that information. This limited the effectiveness of advertising and made it difficult to improve user engagement.

[1222] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1223] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for analyzing the content of the video and identifying scenes and topics, means for selecting related advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for naturally splicing the generated advertising video into the original video, means for delivering the modified video to the user terminal, means for collecting and analyzing user emotion data in real time, and means for reselecting advertising materials based on the user emotion data, thereby enabling advertising delivery optimized to the user's emotions.

[1224] A "user terminal" is a device used by a user to upload and play videos, and includes smartphones, tablets, personal computers, etc.

[1225] A "server" is a computer system that is responsible for receiving, storing, analyzing, generating and distributing advertising materials, and other processes for video files.

[1226] A "database" is a system for systematically storing and managing data such as video files, advertising materials, and analysis results.

[1227] A "video file" is a file containing video and audio data uploaded by a user.

[1228] "Analysis" is the process of breaking down video content frame by frame, identifying its content using image and voice recognition technology, and identifying scenes and topics.

[1229] A "scene" refers to a specific situation or situation within a video, and is a part of a video that consists of elements such as people, objects, and background.

[1230] "Topics" refer to the themes or topics covered in the video, such as introductions to tourist spots or product reviews.

[1231] "Advertising materials" refers to elements such as images, videos, and audio used for advertising distribution, and are promotional content provided by companies.

[1232] "Automatic generation" refers to the process by which a system programmatically generates advertising videos without manual intervention.

[1233] A "modified video" is a video file that has been edited by adding advertising material to the original video.

[1234] "Emotion data" refers to data relating to the user's emotional state extracted from facial expressions, tone of voice, eye movements, and the like.

[1235] "Collection" refers to the process of acquiring emotion data and viewing data from user terminals and storing them on a server.

[1236] "Analysis" is the process of identifying the user's emotional state based on collected emotional data and selecting advertising materials accordingly.

[1237] "Viewing data" refers to information about the user's actions when watching a video, including viewing time, number of views, clicks, and the like.

[1238] "Reward" refers to compensation such as electronic money or points provided to content creators based on viewing data and emotional data.

[1239] The present invention is a system that automatically generates advertisements appropriate for the situation for videos provided by users, recognizes the user's emotions, and utilizes the data. This system is composed of a user terminal, a server, an emotion engine, and a database.

[1240] First, a user uploads a video file through a dedicated website or application. The device used can be a smartphone, tablet, or personal computer. The video file is then received by a server and stored in a database. Specifically, the database may use SQL Server or MySQL.

[1241] The server then breaks down the stored video into frames and extracts the image data for each frame. It uses image recognition algorithms such as OpenCV and speech recognition algorithms such as the Google Speech-to-Text API. The server then uses these algorithms to identify scenes and topics in the video.

[1242] Based on the identified scenes and topics, the server selects relevant advertising materials from the advertising provider's database. Selection criteria include relevance to the scene or topic, and the preferences of the target audience. The server then combines the selected advertising materials and automatically generates an advertising video using a video editing algorithm (e.g., FFmpeg).

[1243] After generating the ad video, the server uses an emotion engine to collect video and audio data from the user's device in real time to obtain the user's emotional data. Microsoft Azure Cognitive Services is used to analyze the emotional data. By analyzing the collected emotional data, the server can understand the user's emotional state while watching and can reselect advertising materials based on that data.

[1244] The modified video file is then delivered to the user's device by the server. This delivery utilizes streaming technology, such as RTMP (Real-Time Message Protocol). Metadata is added to the modified video, and user viewing status is tracked. The user then plays the modified video on their device and begins watching.

[1245] Finally, the server compiles ad viewing rates and effectiveness indicators based on the collected viewing and emotion data. Based on the compilation results, rewards are provided to content creators using the viewing and emotion data. Rewards are credited to the content creators' accounts as electronic money or points.

[1246] As a concrete example, consider the case where a user uploads a travel vlog video. The server analyzes the video and identifies scenes such as "introducing tourist spots" and "tasting local cuisine." Next, it selects advertising materials related to these scenes and seamlessly stitches the automatically generated advertising videos into the original video. When a user plays this modified video, the server obtains emotional data in real time and delivers optimal advertising, thereby improving user engagement.

[1247] Example prompt: "Describe how an AI system can automatically generate relevant ads for travel vlogs that showcase tourist attractions when uploaded."

[1248] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1249] Step 1:

[1250] A user uploads a video.

[1251] Input: A video file selected by the user.

[1252] Output: The video file that will be sent to the server.

[1253] How it works: The user selects a video file through a dedicated website or application and clicks the upload button. The video file is then sent to the server via the Internet.

[1254] Step 2:

[1255] The server receives the video and stores it in a database.

[1256] Input: The video file submitted by the user.

[1257] Output: Video files stored in the database.

[1258] Specific operation: The server receives the video file and saves it in a database (e.g., MySQL or SQL Server). When saving, the video file's metadata (e.g., file name, upload date, etc.) is also registered.

[1259] Step 3:

[1260] The server analyzes the video.

[1261] Input: Video files stored in the database.

[1262] Output: Image data decomposed into frames, and audio data converted to text.

[1263] What it does: The server breaks down the video file into frames, extracts the image data for each frame using OpenCV, and converts the audio data to text using the Google Speech-to-Text API.

[1264] Step 4:

[1265] The server identifies the scene and topic.

[1266] Input: Image data decomposed frame by frame and audio data converted to text.

[1267] Output: Data about identified scenes and topics.

[1268] How it works: The server uses image recognition and text analysis algorithms to identify scenes (e.g., introductions to tourist attractions, tastings of local cuisine) and topics within the video.

[1269] Step 5:

[1270] The server selects the advertising material.

[1271] Input: Data about the identified scene and topic.

[1272] Output: Selected advertising material data.

[1273] Specific operation: Based on the analysis results, the server issues a query to the ad provider's database to obtain relevant ad material data. Selection criteria take into account the relevance to the scene or topic, the preferences of the target audience, etc.

[1274] Step 6:

[1275] The server generates the advertisement video.

[1276] Input: Selected advertising material data.

[1277] Output: Auto-generated ad video.

[1278] Specific operation: The server combines the advertising materials and automatically generates an advertising video using a video editing algorithm (e.g., FFmpeg). The generated advertising video is edited to seamlessly connect with the original video.

[1279] Step 7:

[1280] The server collects and analyzes the emotion data.

[1281] Input: Real-time video and audio data sent from the user terminal.

[1282] Output: Parsed emotion data.

[1283] How it works: The server uses an emotion engine (such as Microsoft Azure Cognitive Services) to collect and analyze the user's video and audio data in real time. The analysis results include the user's facial expressions, tone of voice, and eye movements.

[1284] Step 8:

[1285] The server delivers the video to the user's device.

[1286] Input: The modified video file.

[1287] Output: A streaming URL that can be played on the user's device.

[1288] Specific operation: The server uploads the modified video file to the streaming service, generates a streaming URL, and notifies the user of this URL.

[1289] Step 9:

[1290] The server collects viewing data and provides rewards.

[1291] Input: User viewing and emotion data.

[1292] Output: Aggregated advertising effectiveness metrics and rewards for content creators.

[1293] Specific operation: The server monitors video playback status and emotional data in real time, aggregates viewing data, and awards electronic money or points to content creators through a reward system based on the aggregated results.

[1294] (Application example 2)

[1295] 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."

[1296] Conventional advertising systems lack the ability to automatically select and generate effective advertising materials for video content provided by users, making it difficult to deliver ads that take into account user emotions and the content of scenes. Furthermore, technology for optimizing ads based on viewer emotions is immature, and compensation for content creators is limited to viewing data. The purpose of this invention is to solve these issues and maximize advertising effectiveness while providing excellent convenience for viewers and content creators.

[1297] 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.

[1298] In this invention, the server includes means for receiving video files from a user terminal and storing them in a database, means for analyzing the content of the video and identifying scenes and topics, means for selecting related advertising materials based on the identified scenes and topics, means for automatically generating an advertising video by combining the selected advertising materials, means for naturally splicing the generated advertising video into the original video, means for delivering the modified video to the user terminal, and means for analyzing user emotions in real time and optimizing the advertising materials using the data. This makes it possible to analyze user emotions and video scenes in real time and automatically generate and deliver optimized advertising videos.

[1299] A "user terminal" is a device used by a user, such as a computer, smartphone, or tablet.

[1300] A "video file" is a digital file containing video and audio provided by a user.

[1301] A "database" is a system for storing and managing video files and related information.

[1302] The "server" is a centralized management system for managing the database, analyzing videos, and generating advertisements.

[1303] A "scene" refers to a specific situation or situation within a video.

[1304] "Topics" are themes or topics covered in the video.

[1305] "Advertising materials" are content such as video, audio, and text that are the source of the advertisement to be distributed.

[1306] An "advertising video" is an advertising video inserted into a video provided by a user.

[1307] "Emotion" refers to the emotional response or state that a user shows when watching a video.

[1308] "Advertising optimization" is the act of selecting and editing optimal advertising materials based on user emotions and video scenes.

[1309] "Viewing data" is data that records the actions and reactions of users when they watch videos.

[1310] "Reward" refers to monetary or points provided to content creators based on viewing data and emotional data.

[1311] To implement this invention, several major pieces of hardware and software are required, such as a server, a user terminal, a database, and an emotion engine.

[1312] First, a user shoots a video file using their own device and uploads it to the server via a dedicated application or website. User devices can be PCs, smartphones, tablets, etc. The server receives the uploaded video file and stores it in a database. The database centrally manages video files and their metadata.

[1313] The server then breaks down the saved video file into frames and analyzes the image and audio data of each frame. This process utilizes image and audio recognition algorithms. Specifically, OpenCV is used to analyze the image frames, and DeepFace and other algorithms are used to recognize facial expressions. Natural language processing libraries (such as the Hugging Face transformer) are used for audio recognition.

[1314] The server analyzes the content of the video to identify its scenes and topics, then selects advertising materials appropriate for the scenes from a related database, taking into account the content of the scenes and the attributes of the target audience.

[1315] Once the ad material selection is complete, the server automatically generates an ad video using the selected ad material. At this time, a video editing library (e.g., MoviePy) is used to seamlessly stitch the ad video into the original video. The generated ad video is designed to maintain the flow of the original video.

[1316] The modified video is then sent back to the user's device from the server. The user watches the modified video, and during this process the server analyzes the user's emotions in real time. DeepFace and other emotion analysis engines are used as emotion engines. For example, the server collects user emotional data based on eye movements, facial expressions, and tone of voice.

[1317] Furthermore, the system tracks the viewing status of the modified videos and aggregates the viewing data. This viewing data and emotional data are used to provide compensation to content creators. The compensation is credited to the content creator's account in the form of electronic money or points.

[1318] For example, suppose a user uploads a travel vlog video. In this case, the server analyzes the video and identifies scenes such as "introducing tourist spots" and "tasting local cuisine." It then selects advertising materials for sightseeing tours that are suitable for the "introducing tourist spots" scene and automatically generates a 15-second advertising video based on them. The generated advertising video is then seamlessly integrated into the original video and optimized based on the user's emotional data.

[1319] An example of a prompt sentence is as follows:

[1320] Generate a Python program that performs frame-based emotion recognition and inserts appropriate ads.

[1321] By using such prompts, it is possible to have a generative AI model automatically generate a program to perform a specific task.

[1322] This makes it possible to analyze user emotions and video scenes in real time, and automatically generate and deliver optimized advertising videos.

[1323] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1324] Step 1:

[1325] Upload video files from the user's device

[1326] Users use their own devices to upload video files to the server through a dedicated application or website. At this time, the video files are stored in a database. The input is the video file recorded by the user, and the output is the video data stored in the database.

[1327] Step 2:

[1328] Video decomposition and frame analysis

[1329] The server breaks down the stored video file into frames. It uses OpenCV to analyze the image data of each frame and recognizes faces, objects, etc. The input is the video file stored in the database, and the output is the image data of each frame. Specifically, it uses an image recognition algorithm to analyze people, backgrounds, etc. in the video.

[1330] Step 3:

[1331] Audio analysis

[1332] The server uses a speech recognition algorithm to analyze the audio data in the video file, convert it into text, and identify the scenes and topics in the video. The input is the audio data for each frame, and the output is text data.

[1333] Step 4:

[1334] Identifying scenes and topics

[1335] The server integrates image data and audio data to identify scenes and topics within the video. Natural language processing algorithms and emotion recognition models are used here. The input is image data and audio text data for each frame, and the output is information on the identified scenes and topics.

[1336] Step 5:

[1337] Selection of advertising materials

[1338] The server selects the most suitable advertising material from a database based on the identified scene and topic, taking into account factors such as the relevance to the scene and the preferences of the target audience. The input is information about the scene and topic, and the output is the selected advertising material.

[1339] Step 6:

[1340] Ad video generation

[1341] The server automatically generates an ad video using the selected ad material. Using a video editing library such as MoviePy, the ad video is seamlessly spliced ​​into the original video. The input is the original video frames and ad material, and the output is the generated ad video.

[1342] Step 7:

[1343] Distribution of altered videos

[1344] The server integrates the generated advertising video with the original video and delivers the modified video file to the user terminal. The input is the generated advertising video and the original video, and the output is the modified video file.

[1345] Step 8:

[1346] User sentiment analysis

[1347] When a user watches the modified video, the server uses an emotion engine to analyze the user's reactions (facial expressions, gaze, tone of voice, etc.). The input is the video and audio data the user is watching, and the output is the user's emotional data.

[1348] Step 9:

[1349] Tracking and aggregating viewing data

[1350] The server tracks the viewing status of the modified video and aggregates the viewing data. The input is user viewing behavior data, and the output is aggregated viewing data.

[1351] Step 10:

[1352] Providing rewards

[1353] The server provides rewards to content creators based on the viewing and emotion data. The rewards are added to their accounts as electronic money or points. The input is the aggregated viewing and emotion data, and the output is the rewards to the content creators.

[1354] 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.

[1355] 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.

[1356] 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.

[1357] 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.

[1358] 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.

[1359] 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.

[1360] 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).

[1361] 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.

[1362] 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."

[1363] 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.

[1364] 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).

[1365] 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.

[1366] 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.

[1367] 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.

[1368] 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.

[1369] 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.

[1370] 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.

[1371] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1372] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1373] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1374] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1375] The following is further disclosed regarding the above embodiment.

[1376] (Claim 1)

[1377] A means for receiving video files from a user terminal and storing them in a database;

[1378] A means of analyzing the content of a video and identifying scenes and topics,

[1379] a means for selecting relevant advertising materials based on the identified scenes or topics;

[1380] A means to automatically generate advertising videos by combining selected advertising materials,

[1381] A means to naturally connect the generated advertising video to the original video,

[1382] and means for delivering the modified video to a user terminal.

[1383] (Claim 2)

[1384] 10. The system of claim 1, further comprising means for tracking viewing of the modified video and aggregating viewing data.

[1385] (Claim 3)

[1386] 10. The system of claim 1, further comprising means for providing compensation to content creators based on the viewing data.

[1387] "Example 1"

[1388] (Claim 1)

[1389] A means for receiving video files from a user terminal and storing them in a database;

[1390] A means for decomposing a video into frames and extracting image data for each frame;

[1391] A means for analyzing the extracted image data to identify the content of people, objects, background, etc.;

[1392] A means of converting audio data into text to identify scenes and topics,

[1393] a means for selecting relevant advertising materials based on the identified scenes or topics;

[1394] A means to automatically generate advertising videos by combining selected advertising materials,

[1395] A means to naturally connect the generated advertising video to the original video,

[1396] and means for delivering the modified video to a user terminal.

[1397] (Claim 2)

[1398] 10. The system of claim 1, further comprising means for tracking viewing of the modified video and aggregating viewing data.

[1399] (Claim 3)

[1400] 10. The system of claim 1, further comprising means for providing compensation to content creators based on the viewing data.

[1401] "Application Example 1"

[1402] (Claim 1)

[1403] A means for receiving video files from a user terminal and storing them in a database;

[1404] A means of analyzing the content of a video and identifying scenes and topics,

[1405] a means for selecting relevant advertising materials based on the identified scenes or topics;

[1406] A means to automatically generate advertising videos by combining selected advertising materials,

[1407] A means to naturally connect the generated advertising video to the original video,

[1408] means for delivering the modified video to a user terminal;

[1409] A means of tracking viewing data in real time; and

[1410] means for rearranging advertising videos based on viewing data;

[1411] A system including:

[1412] (Claim 2)

[1413] 10. The system of claim 1, further comprising means for tracking viewing of the modified video and aggregating viewing data.

[1414] (Claim 3)

[1415] 10. The system of claim 1, further comprising means for providing compensation to content creators based on the viewing data.

[1416] "Example 2: Combining Emotion Engines"

[1417] (Claim 1)

[1418] A means for receiving video files from a user terminal and storing them in a database;

[1419] A means of analyzing the content of a video and identifying scenes and topics,

[1420] a means for selecting relevant advertising materials based on the identified scenes or topics;

[1421] A means to automatically generate advertising videos by combining selected advertising materials,

[1422] A means to naturally connect the generated advertising video to the original video,

[1423] means for delivering the modified video to a user terminal;

[1424] means for collecting and analyzing user emotion data in real time;

[1425] A system including means for reselecting advertising materials based on user emotion data.

[1426] (Claim 2)

[1427] Tracking viewing behavior and real-time sentiment data for the altered videos,

[1428] 10. The system of claim 1, further comprising means for aggregating the viewing data and the emotional data.

[1429] (Claim 3)

[1430] and means for providing rewards to content creators based on the viewing data and the sentiment data.

[1431] 10. The system of claim 1.

[1432] "Application example 2 when combining emotion engines"

[1433] (Claim 1)

[1434] A means for receiving video files from a user terminal and storing them in a database;

[1435] A means of analyzing the content of a video and identifying scenes and topics,

[1436] a means for selecting relevant advertising materials based on the identified scenes or topics;

[1437] A means to automatically generate advertising videos by combining selected advertising materials,

[1438] A means to naturally connect the generated advertising video to the original video,

[1439] means for delivering the modified video to a user terminal;

[1440] A system that analyzes user emotions in real time and uses that data to optimize advertising materials.

[1441] (Claim 2)

[1442] 10. The system of claim 1, further comprising means for tracking viewing of the modified video and aggregating viewing data.

[1443] (Claim 3)

[1444] 10. The system of claim 1, further comprising means for providing rewards to content creators based on the viewing data and the emotional data. [Explanation of symbols]

[1445] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for receiving video files from a user terminal and storing them in a database; A means of analyzing the content of a video and identifying scenes and topics, a means for selecting relevant advertising materials based on the identified scenes or topics; A means to automatically generate advertising videos by combining selected advertising materials, A means to naturally connect the generated advertising video to the original video, and means for delivering the modified video to a user terminal.

2. The system of claim 1 , further comprising means for tracking viewing of the modified video and aggregating viewing data.

3. The system of claim 1 , further comprising means for providing compensation to content creators based on the viewing data.

Images