Information recommendation processing method, device and equipment and computer readable storage medium

Abstract

The application provides an information recommendation processing method and device, electronic equipment and a computer readable storage medium. The method comprises: displaying an information stream page; in response to an image recognition trigger operation received in the information stream page, displaying a captured image; when the captured image comprises an interactive object and the interactive object is associated with an item to be recommended in the information stream page, presenting an interactive result related to the item to be recommended. Through the application, the recommendation efficiency of the item to be recommended can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the Internet technology, and particularly to an information recommendation processing method and device, an electronic device, and a computer readable storage medium. BACKGROUND

[0002] With the development of Internet technology, inserting recommended information in an information flow page can not only realize diversified information display of the information flow, but also realize recommended information diversion. In related technologies, the display mode of the recommended information is single and fixed, and only one-sided information exposure is realized, which is difficult to interact with users, leading to ineffective recommendation and wasting server resources, and affecting the recommendation efficiency of the recommended information. SUMMARY

[0003] The embodiments of the present application provide an information recommendation processing method, device, electronic device, and computer readable storage medium, which can realize accurate information recommendation.

[0004] The technical solutions of the embodiments of the present application are implemented as follows:

[0005] The embodiments of the present application provide an information recommendation processing method, which comprises the following steps:

[0006] displaying an information flow page;

[0007] displaying a collected image in response to an image recognition trigger operation received in the information flow page;

[0008] when the collected image includes an interactive object and the interactive object is associated with an item to be recommended in the information flow page, presenting an interactive result related to the item to be recommended.

[0009] In the above solution, the method further comprises the following steps:

[0010] sending a sharing message corresponding to the interactive object in response to a sharing operation on the interactive result;

[0011] The sharing message is used to jump to the image recognition portal when triggered.

[0012] In the above solution, the presenting the interactive result related to the item to be recommended comprises the following steps:

[0013] determining the difference between the interactive object recognized from the collected image and an interactive object icon, determining the score of the collected image according to the difference, and presenting an interactive result associated with the score;

[0014] The difference includes at least one of the following:

[0015] a position difference between the recognized interactive object and the interactive object illustration;

[0016] a state difference between the recognized interactive object and the interactive object illustration;

[0017] an image quality difference between the recognized interactive object and the interactive object illustration.

[0018] In the above scheme, the presenting the interactive result related to the to-be-recommended item comprises:

[0019] determining a score of each of the collected images based on a difference between the recognized interactive object and the interactive object illustration in each of the collected images;

[0020] determining a number of images whose scores exceed a score threshold, and presenting an interactive result associated with the number.

[0021] Embodiments of the present application provide an information recommendation processing apparatus, comprising:

[0022] a display module configured to display an information stream page;

[0023] a trigger module configured to display a collected image in response to an image recognition trigger operation received in the information stream page;

[0024] an interactive module configured to present an interactive result related to a to-be-recommended item when the collected image includes an interactive object and the interactive object is associated with the to-be-recommended item in the information stream page.

[0025] In the above scheme, the display module is further configured to:

[0026] display a first-level information stream page;

[0027] The first-level information stream page is a default information stream page displayed when a client is started, and the information stream in the first-level information stream page includes at least one of the following: at least one historical session in which a login account participates; a subscription message of the login account; and a notification message of the login account.

[0028] In the above scheme, the display module is further configured to:

[0029] display an image recognition entry in the first-level information stream page;

[0030] When a trigger operation received for the image recognition entry is determined to be the image recognition trigger operation.

[0031] In the above scheme, the display module is further configured to:

[0032] display a second-level information stream page;

[0033] The second-level information stream page includes social dynamics of a login account in an information stream.

[0034] In the above solution, the display module is further configured to:

[0035] insert and display at least one recommended information in the information stream of the second-level information stream page;

[0036] determine the received trigger operation on the recommended information as the image recognition trigger operation.

[0037] In the above solution, the display module is further configured to:

[0038] insert and display at least one recommended information in the information stream of the second-level information stream page;

[0039] present a detail page of the recommended information in response to a trigger operation on the recommended information, and display an image recognition entry in the detail page;

[0040] determine the received trigger operation on the image recognition entry as the image recognition trigger operation.

[0041] In the above solution, the apparatus further includes a sharing module configured to:

[0042] send a sharing message corresponding to the interactive object in response to a sharing operation on the interactive result;

[0043] The sharing message is configured to jump to the image recognition entry when triggered.

[0044] In the above solution, when the image recognition entry is displayed, the display module is further configured to:

[0045] display first prompt information to prompt the interactive object to be collected;

[0046] The first prompt information includes at least one of the following: introduction information of the interactive object, and a diagram of the interactive object.

[0047] In the above solution, the display module is further configured to:

[0048] present second prompt information to prompt to continue image collection when the collected image does not include the interactive object associated with the item to be recommended in the information stream page;

[0049] present a special effect corresponding to the interactive object when the collected image includes the interactive object associated with the item to be recommended;

[0050] The special effect includes at least one of the following:

[0051] Animation, prompt tone, vibration feedback.

[0052] In the above scheme, the interaction module is further configured to:

[0053] Identify the state of the interactive object, and present an interaction result associated with the state of the interactive object;

[0054] The interactive object includes at least one of the following: limbs, hands, and faces.

[0055] The state of the interactive object includes at least one of the following: limb movements, hand gestures, and facial expressions.

[0056] The interaction result includes at least one of the following: electronic red packets, electronic exchange coupons, and electronic discount coupons.

[0057] In the above scheme, the interaction module is further configured to:

[0058] Determine an effective image for target recognition from the collected image;

[0059] Invoke an object recognition model to perform target recognition on the effective image to determine the type of object included in the effective image;

[0060] Invoke a state type recognition model to perform key point positioning processing on the object included in the effective image to obtain the state of the object.

[0061] In the above scheme, the interaction module is further configured to:

[0062] When the collected image is a static image collected by taking a photo, determine the static image as an effective image for target recognition;

[0063] When the collected image is a dynamic video frame collected by taking a video, determine a video frame that meets at least one of the following conditions from a plurality of video frames as an effective image for target recognition:

[0064] The average frame difference between the video frame and an adjacent video frame is less than a still frame threshold;

[0065] The variance of the first-order gradient of the video frame is greater than a definition threshold.

[0066] In the above scheme, the interaction module is further configured to, before determining an effective image for target recognition from the collected image:

[0067] Obtain a recommended validity period;

[0068] When the acquisition time of the acquired image is within the recommended validity period, it is determined that the operation of determining valid images for target recognition from the acquired images will be performed.

[0069] In the above scheme, the interaction module is further configured to:

[0070] determine a difference between the recognized interaction object and the interaction object illustration, determine a score of the acquired image according to the difference, and present an interaction result associated with the score;

[0071] The difference includes at least one of the following:

[0072] a position difference between the recognized interaction object and the interaction object illustration;

[0073] a state difference between the recognized interaction object and the interaction object illustration;

[0074] an image quality difference between the recognized interaction object and the interaction object illustration.

[0075] In the above scheme, the interaction module is further configured to:

[0076] determine a score of each of the acquired images based on the difference between the recognized interaction object and the interaction object illustration in each of the acquired images;

[0077] determine the number of images whose scores exceed a score threshold, and present an interaction result associated with the number.

[0078] Embodiments of the present application provide an electronic device, comprising:

[0079] a memory configured to store executable instructions;

[0080] a processor configured to execute the executable instructions stored in the memory, and implement the information recommendation processing method provided by embodiments of the present application.

[0081] Embodiments of the present application provide a computer readable storage medium storing executable instructions, which are configured to cause a processor to execute the information recommendation processing method provided by embodiments of the present application.

[0082] Embodiments of the present application have the following beneficial effects:

[0083] By fusing the interactive recommended items in the information flow page, the recommended information can accurately hit the demand of the user in the information flow browsing scene, and the interactive results related to the recommended items are presented in the man-machine interactive mode, which can not only enrich the presentation mode of the recommended items to improve the recommendation efficiency of the recommended items, but also provide diversified user interactive experience. BRIEF DESCRIPTION OF DRAWINGS

[0084] Figure 1 is a structural schematic diagram of an information recommendation processing system architecture provided by an embodiment of the present application;

[0085] Figure 2 is a structural schematic diagram of a terminal of an information recommendation processing method provided by an embodiment of the present application;

[0086] Figures 3A-3C is a flow schematic diagram of an information recommendation processing method provided by an embodiment of the present application;

[0087] Figures 4A-4L is an interface schematic diagram of an information recommendation processing method provided by an embodiment of the present application;

[0088] Figure 5 is an identification flow schematic diagram of an information recommendation processing method provided by an embodiment of the present application;

[0089] Figures 6A-6B is a model schematic diagram of an information recommendation processing method provided by an embodiment of the present application;

[0090] Figure 7 is an identification principle schematic diagram of an information recommendation processing method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0091] In order to make the objectives, technical solutions and advantages of the present application clearer, the following will further describe the present application in conjunction with the accompanying drawings, and the described embodiments should not be regarded as limiting the present application, and all other embodiments obtained by those skilled in the art without creative labor shall fall within the scope of protection of the present application.

[0092] In the following description, "some embodiments" are described, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict.

[0093] In the following description, the terms "first\second" are merely to distinguish similar objects, and do not represent a specific order of the objects, and it can be understood that "first\second" can be interchanged in a specific order or sequence as allowed, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0094] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description herein is for describing particular embodiments only and is not intended to be limiting of the application.

[0095] The relevant data collection processing in the embodiments of the application should strictly comply with the requirements of relevant laws and regulations, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing within the scope of authorization of laws and regulations and the personal information subject.

[0096] Before the embodiments of the application are further described in detail, the terms and phrases involved in the embodiments of the application are explained, and the terms and phrases involved in the embodiments of the application are applicable to the following explanations.

[0097] 1) Information flow (Feeds) advertising: advertising displayed in the social media user friend dynamic or information media and audiovisual media content flow;

[0098] 2) Single-shot detector (SSD): a feature pyramid structure is used for detection, that is, different feature maps are used for classification and position regression at the same time.

[0099] The interaction form of the recommended item in the related art is usually in the form of text, button, card, picture, video, etc. to attract the user to click, and after receiving the click operation of the user, the recommended item advertisement page is entered. The applicant found in the implementation of the embodiments of the application that the recommendation method of the recommended item in the related art lacks click motivation for the user, thereby affecting the conversion of the advertisement. Even after entering the recommended item advertisement page after receiving the click operation of the user, the user stays in the recommendation page for too short a time to ensure that the information of the recommended item is effectively presented to the user (the user stays in the recommendation page for more than a stay time threshold to represent that the information is effectively presented to the user), thereby affecting the conversion of the advertisement, and causing waste of resources for pushing the advertisement.

[0100] The embodiments of the application provide an information recommendation processing method and device, electronic equipment and computer readable storage medium, which can improve the recommendation efficiency of the recommended item. The following describes an exemplary application of the electronic equipment provided by the embodiments of the application. The device provided by the embodiments of the application can be implemented as a notebook computer, a tablet computer, a desktop computer, a set-top box, a mobile device (for example, a mobile phone, a portable music player, a personal digital assistant, a dedicated message device, a portable game device) and various types of user terminals. It can also be implemented as a server. In the following, an exemplary application of the device implemented as a terminal will be described.

[0101] Referring to Figure 1 , Figure 1 is a structural schematic diagram of an information recommendation processing system architecture provided by an embodiment of the present application, Figure 1 An information recommendation processing system 100 is shown, a terminal 400 connects a delivery server 200-1 and a computing server 200-2 through a network 300-1, and the network 300-1 can be a wide area network or a local area network, or a combination of the two. According to the different delivery content, the delivery server can be an advertisement server for delivering advertisements, and the delivery server can also be a news server for delivering news.

[0102] The delivery server 200-1 delivers a detection task and an advertisement configuration to the terminal 400, the terminal 400 presents an information stream, and presents an image recognition portal in the information stream, and when an image recognition trigger operation is detected, an image is collected and sent to the computing server 200-2 for target recognition, and when the identified result is consistent with the advertisement configuration, the identified result is returned to the terminal 400, so that the terminal 400 presents the advertisement information corresponding to the identified result.

[0103] Referring to Figure 1 , both the server and the terminal can join the blockchain network 300-2 and become a node in the blockchain network 300-2. The type of blockchain network 300-2 is flexible and diverse, for example, it can be any one of a public chain, a private chain or an alliance chain. Taking a public chain as an example, any business subject electronic device such as a terminal can access the blockchain network 300-2 without authorization to serve as a consensus node of the blockchain network 300-2, for example, the delivery server 200-1 is mapped to the consensus node 300-1 in the blockchain network 300-2, the computing server 200-2 is mapped to the consensus node 300-2 in the blockchain network 300-2, and the terminal 400 is mapped to the consensus node 300-0 in the blockchain network 300-2.

[0104] Taking the blockchain network 300-2 as an example of a consortium chain, the server and the terminal can access the blockchain network only after the electronic devices under the server and the terminal obtain authorization. The client in the terminal 400 receives the initiator's request for obtaining an interactive object, and the computing server 200-2 sends the proposal for determining the interactive object to the delivery server 200-1 and other terminals. The delivery server 200-1 and other terminals can verify the proposal for determining the interactive result by executing the smart contract. When a number threshold of nodes confirm that the verification is passed, the verification manner is to query whether the terminal 400 has the permission to obtain the interactive result in the ledger of the blockchain network, and whether there is a stock of interactive results of the corresponding category. Each delivery server 200-1 and other terminals verify and sign a digital signature (i.e., endorsement) after verification. When a proposal for determining the interactive result has sufficient endorsements, the jump page address of the interactive result is determined, and the jump page address of the interactive result is returned to the terminal to present the interactive result. By consensus verification of multiple nodes on the virtual resource proposal to be presented, the verification overhead of the server on the proposal for determining the interactive result to be presented can be saved. By verification of multiple terminals on the proposal for determining the interactive result, since the virtual resource is encapsulated in the interactive result, the risk of repeated delivery or excessive delivery of the interactive result can be reduced by ensuring the reliability of the proposal.

[0105] In some embodiments, the terminal implements the session processing method provided in the embodiments of the present application by running a computer program. The computer program can be a native program or a software module in the operating system; can be a native application program (APP, Application), that is, a program that needs to be installed in the operating system to run; can also be a small program, that is, a program that only needs to be downloaded into a browser environment to run; and can also be an instant messaging small program or a file management small program that can be embedded into any APP. In summary, the above computer program can be any application program, module or plug-in in any form.

[0106] In some embodiments, the server can be a standalone physical server, a server cluster or a distributed system composed of multiple physical servers, a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDNs, and basic cloud computing services such as big data and artificial intelligence platforms. The terminal 400 can be a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. The terminal and the server can be directly or indirectly connected through wired or wireless communication, which is not limited in the embodiments of the present application.

[0107] Referring to Figure 2 ,Figure 2 is a structural schematic diagram of a terminal of an application information recommendation method provided by an embodiment of the present application, Figure 2 The terminal 400 shown includes at least one processor 410, a memory 450, at least one network interface 420, and a user interface 430. The various components in the terminal 400 are coupled together by a bus system 440. It can be understood that the bus system 440 is used to realize the connection communication between the components. In addition to including a data bus, the bus system 440 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, all the buses are marked as the bus system 440 in Figure 2 .

[0108] The processor 410 can be an integrated circuit chip having a processing capability of a signal, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., wherein the general-purpose processor can be a microprocessor or any conventional processor.

[0109] The user interface 430 includes one or more output devices 431 that enable display of media content, including one or more speakers and / or one or more visual display screens. The user interface 430 also includes one or more input devices 432 that facilitate user input, such as a keyboard, a mouse, a microphone, a touch screen display, a camera, other input buttons and controls.

[0110] The memory 450 can be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state memory, hard drives, optical drives, and the like. The memory 450 optionally includes one or more storage devices physically located in proximity to the processor 410.

[0111] The memory 450 includes volatile memory or non-volatile memory, and can also include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), and the volatile memory can be random access memory (RAM). The memory 450 described in the embodiments of the present application is intended to include any suitable type of memory.

[0112] In some embodiments, the memory 450 is capable of storing data to support various operations, examples of which include programs, modules, and data structures or subsets or supersets thereof, which are exemplarily illustrated below.

[0113] The operating system 451 includes system programs for processing various basic system services and performing hardware-related tasks, such as a framework layer, a core library layer, a driver layer, and the like, for implementing various basic services and processing hardware-based tasks;

[0114] The network communication module 452 is configured to reach other computing devices via one or more (wired or wireless) network interfaces 420, exemplary network interfaces 420 including Bluetooth, wireless compatibility authentication (WiFi), and universal serial bus (USB), and the like;

[0115] The display module 453 is configured to enable display of information (e.g., a user interface for operating peripheral devices and displaying content and information) via one or more output devices 431 (e.g., a display screen, a speaker, and the like) associated with the user interface 430.

[0116] The input processing module 454 is configured to detect and interpret one or more user inputs or interactions from one or more input devices 432.

[0117] In some embodiments, the information recommendation processing apparatus provided by the embodiments of the present application can be implemented in a software manner, Figure 2 An information recommendation processing apparatus 455 stored in the memory 450 is shown, which can be software in the form of programs and plug-ins, including the following software modules: a display module 4551, a triggering module 4552, an interaction module 4553, and a sharing module 4554. These modules are logical, and thus can be combined or further split according to the implemented functions. The functions of each module will be described below.

[0118] The information recommendation processing method provided by the embodiments of the present application will be described in conjunction with exemplary applications and implementations of the terminal provided by the embodiments of the present application. The execution subject of the following method is the terminal, which can be specifically implemented by the terminal running the various computer programs described above. Of course, according to the understanding of the following, it is not difficult to see that the terminal and the server can also cooperate to implement the information recommendation processing method provided by the embodiments of the present application.

[0119] Referring to Figure 3A , Figure 3A is a flowchart of the information recommendation processing method provided by the embodiments of the present application, which will be described in conjunction with the steps shown in Figure 3A .

[0120] In step 101, an information stream page is displayed.

[0121] In some embodiments, the step 101 of displaying the information stream page can be implemented by the following technical solution: displaying a first-level information stream page; wherein the first-level information stream page is an information stream page displayed by default when the client is started, and the information stream in the first-level information stream page comprises at least one of the following: at least one historical session in which the login account participates; a subscription message of the login account; a notification message of the login account.

[0122] As an example, the information stream page comprises a first-level information stream page and a second-level information stream page, and the first-level information stream page can be a social information stream in the social media client that has a social relationship with the social network account, for example, there are three friends of the login account of Li Si, and the three friends and the login account of Li Si all have historical sessions, so the historical sessions between the three friends and the login account of Li Si are included in the first-level information stream, the login account of Li Si has a subscription message, for example, the login account of Li Si has historical messages of subscribing to a public account, and these historical messages can also be part of the first-level information stream page, and the notification message of the service used by the login account of Li Si can also be part of the first-level information stream page, and these notification messages are services purchased by the login account.

[0123] In some embodiments, when the first-level information stream page is displayed, the following technical solution can also be performed: displaying an image recognition entry in the first-level information stream page; determining that a trigger received for the image recognition entry is an image recognition trigger operation.

[0124] As an example, the image recognition entry in the first-level information stream page can be a functional control, which is suspended in the first-level information stream page, or a top or bottom component of the first-level information stream, and the trigger received by the display driver of the operating system for the image recognition entry is determined as an image recognition trigger operation.

[0125] In some embodiments, the step 101 of displaying the information stream page can be implemented by the following technical solution: displaying a second-level information stream page; wherein the information stream in the second-level information stream page comprises social dynamics of the login account.

[0126] As an example, the information stream in the second-level information stream page comprises social dynamics of the login account, for example, the microblog homepage is a second-level information stream page, and the second-level information stream page presents social dynamics of the login account and social dynamics of accounts having a social relationship with the login account. The second-level information stream page can also present information presented by a specific application client, for example, the second-level information stream page can be a homepage information stream of a news client, comprising a plurality of news information.

[0127] In some embodiments, when the second-level information stream page is displayed, the following technical solutions can also be performed: inserting display of at least one recommended information in the information stream of the second-level information stream page; determining the received trigger operation on the recommended information as an image recognition trigger operation.

[0128] For example, one recommended information is inserted and displayed in the information stream of the second-level information stream page, for example, a recommended information is inserted and displayed in the social dynamics of the logged-in account and the social dynamics of the account having a social relationship with the logged-in account, where the recommended information is neither published by the logged-in account nor published by the account having a social relationship with the logged-in account, for example, a recommended information is inserted and displayed in the information stream (multiple news information) of the home page of the news client, where the recommended information is not news information.

[0129] Referring to Figure 3B , Figure 3B is a flowchart of an information recommendation processing method provided by the embodiments of the present application, which will be described in combination with the steps shown in Figure 3B Step 101 displays an information stream page, which can be implemented through steps 1011-1013.

[0130] In step 1011, at least one recommended information is inserted and displayed in the information stream of the second-level information stream page.

[0131] In step 1012, in response to a trigger operation on the recommended information, a detail page of the recommended information is presented, and an image recognition entry is displayed in the detail page.

[0132] For example, the image recognition entry displayed in the detail page can be implemented through the following solutions: the image recognition entry is displayed in a page suspended in the detail page, where the page suspended in the detail page can be a suspension area control, or the image recognition entry is displayed in a bottom component of the detail page, and at the same time, interactive prompt information is also displayed to prompt the activity details and prompt the user to generate the operation of triggering the image recognition entry.

[0133] In step 1013, the received trigger operation on the image recognition entry is determined as an image recognition trigger operation.

[0134] In some embodiments, when the image recognition entry is displayed, the following technical solutions can also be performed: displaying first prompt information to prompt the interactive object to be collected; where the first prompt information includes at least one of the following: introduction information of the interactive object, a diagram of the interactive object.

[0135] As an example, the introduction information of the interactive object includes a brand name of the recommended item, a type of the interactive object, such as a limb, a hand, a face, and the like, and the type of the interactive object can also be an item, such as a flower, a ticket, a poster, a cola, and the like, and the introduction information can further include a state of the interactive object, such as a limb action, a hand gesture, a facial expression, a placement of the cola, and the like.

[0136] In step 102, in response to the image recognition trigger operation received in the information flow page, the image acquisition interface is called to acquire an image, the acquired image is displayed in a preview page, and target recognition is performed.

[0137] In some embodiments, the following technical solutions can also be performed: determining an effective image for target recognition from the acquired image; calling an object recognition model to perform target recognition on the effective image to determine a type of an object included in the effective image; and calling a state type recognition model to perform key point positioning processing on the object included in the effective image to obtain a state of the object.

[0138] As an example, the object recognition model can be a single point detector, and a BlazeBlock-based basic unit is used to construct a lightweight and efficient backbone network structure based on the single point detector, wherein the multi-scale output of the single point detector is improved to a single-head output, thereby reducing the complexity of processing the maximum threshold and calculation. The state type recognition model is a 21-point three-dimensional skeleton regression model, and the 21-point three-dimensional skeleton regression model is constructed through a similar backbone network structure as the single point detector. While positioning the key points of the regression skeleton, the classification probability of the gesture is output, and the false detection is suppressed while assisting the model learning. Finally, the three-dimensional skeleton coordinates are normalized (mean value reduction and variance division processing) in each dimension to obtain a normalized feature vector, and a three-layer fully connected multilayer perceptron is used to realize the classification of multiple states of each interactive object.

[0139] As an example, the process of calling the object recognition model and the state type recognition model can be implemented on the server side, and the server returns the recognition result, which includes a jump identifier, i.e., jumping to a page of a corresponding interactive result, and the second prompt information can be included in the recognition result to prompt that this recognition fails and image acquisition needs to be performed again.

[0140] In some embodiments, the above-mentioned determining the effective image for target recognition from the collected image can be implemented by the following technical solutions: when the collected image is a static image collected by taking a photo, the static image is determined as the effective image for target recognition; when the collected image is a dynamic video frame collected by taking a video, a video frame meeting at least one of the following conditions among multiple video frames is determined as the effective image for target recognition: the mean value of frame difference between the video frame and an adjacent video frame is less than a static frame threshold; the variance of the first-order gradient of the video frame is greater than a definition threshold.

[0141] As an example, the static image can be a photo, the photo can be a photo taken or the photo can be a stored photo (local storage or remote storage), the static image can be directly used as the effective image for target recognition, and the static image can be further subjected to definition recognition, that is, the variance of the first-order gradient of the static image is calculated, and if the variance of the first-order gradient is greater than a set definition threshold, the static image is clear, that is, the clear static image is used as the effective image for target recognition, otherwise, a prompt information is presented to prompt that the static image does not meet the recognition requirement and needs to be collected again.

[0142] As an example, the collected image can be a dynamic video, that is, the collected image is a plurality of dynamic video frames collected by taking a video, and the effective image for target recognition needs to be obtained from the plurality of video frames, and the clear video frame, the static video frame or the video frame that is both clear and static in all video frames is used as the effective image for target recognition, so as to ensure the accuracy of subsequent object and state recognition, thereby improving the calculation efficiency of the calculation server and reducing the waste of server resources.

[0143] In some embodiments, before determining the effective image for target recognition from the collected image, the following technical solutions can also be performed: obtaining a recommended effective period; when the collection time of the collected image is within the recommended effective period, it is determined that the operation of determining the effective image for target recognition from the collected image will be performed.

[0144] As an example, the recommendation validity period can be obtained from a background server, the recommendation validity period can be a display time of the recommendation information, for example, a certain piece of recommendation information is displayed from 8:00 on January 1, 2020, and will be displayed until 8:00 on January 3, 2020, and the time period during this period is the recommendation validity period, and when the recommendation validity period is within the recommendation validity period, it is determined that the operation of determining the effective image for target recognition from the collected image will be performed, the recommendation validity period can be the effective participation time of the interactive activity in the recommendation information, for example, the recommendation information includes an interactive activity, which starts from 8:00 on January 1, 2020, and will continue until 8:00 on January 3, 2020, and the time period during this period is the recommendation validity period, and when the recommendation validity period is within the recommendation validity period, it is determined that the operation of determining the effective image for target recognition from the collected image will be performed.

[0145] In step 103, when the collected image includes an interactive object, and the interactive object is associated with the item to be recommended in the information stream page, the interactive result related to the item to be recommended is presented.

[0146] As an example, the item to be recommended can be a real product, such as a cosmetic product; the item to be recommended can be a virtual item, such as a game prop; the item to be recommended can be a software product, such as software providing various services.

[0147] In some embodiments, the following technical solutions can also be performed: in response to a sharing operation for the interactive result, a sharing message of the corresponding interactive object is sent; wherein the sharing message is used to jump to the image recognition entrance when triggered.

[0148] As an example, the sharing message includes the collected photo / video and the interactive result, during the image collection process, if a static image is collected, the sharing message includes the collected static image, if a plurality of video frames in a dynamic video are collected, the sharing message includes all video frames, i.e. the dynamic video, the interactive result includes electronic red packets, electronic coupons, and electronic discount coupons, and the sharing message can further include an image recognition entrance link for more users to participate in the interaction through the link, and the link can jump to the image recognition entrance.

[0149] In some embodiments, the following technical solutions can also be performed: when the collected image does not include an interactive object associated with the item to be recommended in the information stream page, a second prompt information is presented to prompt to continue image collection; when the collected image includes an interactive object associated with the item to be recommended, a special effect corresponding to the interactive object is presented; wherein the special effect includes at least one of the following: animation, prompt sound, and vibration feedback.

[0150] As an example, when the collected image does not include the interactive object associated with the item to be recommended in the information flow page, for example, the interactive object is a hand, if the collected image does not include the hand, a second prompt information is directly returned to prompt that the image collection needs to be continued, and when the collected image includes the hand, a special effect corresponding to the interactive object is presented; wherein the special effect includes at least one of the following: animation, prompt sound, vibration feedback, and the special effect can further include a screenshot effect, that is, a special effect of presenting a screenshot of a video frame as an effective image in a static image or dynamic video.

[0151] In some embodiments, the interactive result related to the item to be recommended presented in step 103 can be realized by the following technical solution: identifying the state of the interactive object, and presenting an interactive result associated with the state of the interactive object; wherein the interactive object includes at least one of the following: a limb, a hand, and a face; the state of the interactive object includes at least one of the following: a limb action, a hand gesture, and a facial expression; and the interactive result includes at least one of the following: an electronic red packet, an electronic coupon, and an electronic discount coupon.

[0152] As an example, the interactive result associated with the interactive object or the interactive result associated with the state of the interactive object can be pre-stored in the client or stored in the server, for the former, the server only returns the recognition result (the type of the interactive object, the type and state of the interactive object), and the client queries the corresponding interactive result according to the returned recognition result, that is, jumps to the page of the corresponding interactive result, for the latter, the server not only returns the recognition result (the type of the interactive object, the type and state of the interactive object), but also returns the page address of the corresponding interactive result, and the client directly jumps to the page presenting the interactive result according to the page address.

[0153] As an example, the interactive object can be two hands or two feet, and the two hands are regarded as one interactive object as a whole, that is, there can be multiple objects in each interactive object, and the number of objects is not limited, and learning according to the corresponding training set before applying the model can realize the recognition of the interactive object with two hands and further recognize the state of the two hands, for example, the state of the two hands holding each other is the interactive state of this interactive object (two hands).

[0154] In some embodiments, the interactive result related to the item to be recommended presented in step 103 can be realized by the following technical solution: determining the difference between the interactive object recognized from the collected image and the interactive object diagram, determining the score of the collected image according to the difference, and presenting an interactive result associated with the score; wherein the difference includes at least one of the following: the position difference between the recognized interactive object and the interactive object diagram; the state difference between the recognized interactive object and the interactive object diagram; and the image quality difference between the recognized interactive object and the interactive object diagram.

[0155] In some embodiments, the interaction result related to the recommended item presented in step 103 can be achieved by the following technical solution: determining the score of each collected image based on the difference between the identified interaction object and the interaction object illustration; determining the number of images whose scores exceed the score threshold, and presenting the interaction result associated with the number.

[0156] As an example, the interaction object can be finely divided, and different levels of welfare information can be included in the interaction result, for example, the interaction result includes electronic discount coupons of various discount levels, the collected images are scored according to the difference between the identified interaction object and the interaction object illustration, and the interaction result associated with the score is presented; for example, the position of the identified interaction result matches the interaction object illustration, the state matches (for example, the angle between the gestures matches the angle between the gestures in the illustration), and the image quality also matches (the pixel proximity) The score obtained is higher than the position of the identified interaction result, which does not match the interaction object illustration, the state is more consistent (for example, the model output probability of the identified gesture is greater than the probability threshold of the gesture in the illustration, but it is just reaching the probability threshold, which can be understood as the gesture is recognized but not standard), and the image quality is also inconsistent (pixel proximity, for example, low clarity) The score is higher, and the score is scored according to the three dimensions of position (the closer to the position of the illustration, the higher the position score, and the distance from the position of the illustration can be inversely proportional to the position score), state output probability (the probability value can be directly multiplied to obtain the state score) and clarity (the clarity score can be directly evaluated by the variance of the first-order gradient), each dimension has a corresponding weight, and the final score can be obtained by weighted average according to the corresponding weight, and the recognition result with higher score will correspond to the interaction result with higher discount level.

[0157] As an example, when there are multiple valid images identified, the interaction result can be further divided more finely based on the score and the number when scoring, that is, after determining the score of each collected image according to the above method, determining the number of images whose scores exceed the score threshold, and presenting the interaction result associated with the number, which can include different levels of welfare information, for example, the interaction result includes electronic discount coupons of various discount levels, and the number of valid images whose scores exceed the score threshold is proportional to the discount level of the interaction result, and the higher the number, the greater the discount level.

[0158] Referring to Figure 3C , Figure 3Cis a flowchart of an information recommendation processing method provided by an embodiment of the present application. In step 201, a delivery server issues recommended information, display duration, and interactive image illustrations to a terminal. In step 202, the terminal displays the recommended information in an information stream according to the display duration. In step 203, an image recognition trigger operation is received for the recommended information, and a to-be-recognized image is collected. In step 204, the terminal sends the to-be-recognized image and the interactive image illustrations to a computing server. In step 205, the computing server recognizes the to-be-recognized image. In step 206, when an interactive image matching the interactive image illustrations is recognized, the terminal is returned a recognition result and an interactive result jump identifier corresponding to the recognition result. In step 207, the terminal presents an interactive result according to the received recognition result and the interactive result jump identifier corresponding to the recognition result.

[0159] In the following, an exemplary application of the information recommendation processing method provided by an embodiment of the present application in an actual application scenario will be described.

[0160] In some embodiments, the information recommendation processing method is implemented in an information stream of a social media, as shown in Figure 4A , Figure 4A is an interface diagram of the information recommendation processing method provided by an embodiment of the present application. The information stream of the social media can be a social dynamic information stream. The social dynamic information stream is presented in an information stream display page 401A. The recommended information (advertisement) 402A is presented in the social dynamic information stream. In response to receiving a click operation (image recognition trigger operation) for the recommended information 402A, a guide image 403A is presented. The guide text and background image are presented in the guide image.

[0161] In some embodiments, the information recommendation processing method is implemented in an information stream of a social media, as shown in Figure 4B , Figure 4B is an interface diagram of the information recommendation processing method provided by an embodiment of the present application. The social dynamic information stream is presented in an information stream display page 401B. The recommended information (advertisement) 402B is presented in the social dynamic information stream. In response to receiving a click operation for the recommended information, a detail page 403B of the recommended information is presented. An image recognition entry is presented in a floating area 404B or a bottom component 405B in the detail page. In response to a trigger operation for the image recognition entry, a guide image 406B is presented, and the guide text and background image are presented in the guide image.

[0162] In some embodiments, as shown in Figure 4C , Figure 4Cis a schematic diagram of an interface of the information recommendation processing method provided by the application. When the heart gesture is scanned in the guide picture 401C, the heart-shaped animation 403C appears in the guide picture 401C, and the vibration, the customized green dot pattern, and the screen pause special effect are accompanied. See Figure 4D , Figure 4D is a schematic diagram of an interface of the information recommendation processing method provided by the application. The guide picture 405D (401C in Figure 4C ) includes the display layer 401D that displays page elements, the atmosphere display picture 402D, the circular focus mask layer 403D, and the camera collection display layer 404D. See Figure 4E , Figure 4E is a schematic diagram of an interface of the information recommendation processing method provided by the application. Figure 4E 401E in Figure 4D is a complete display in the atmosphere display picture 402D in Figure 4F , Figure 4F is a schematic diagram of an interface of the information recommendation processing method provided by the application. Figure 4F The guide picture 401F, the guide picture 402F, and the guide picture 403F in

[0163] See Figure 4G , Figure 4G is a schematic diagram of an interface of the information recommendation processing method provided by the application. When the heart gesture is scanned in the guide picture 401G, the half screen 402G is popped up to display the interaction result (the welfare type), but the half screen 402G displays that it is being loaded in the weak network condition.

[0164] See Figure 4H , Figure 4H is a schematic diagram of an interface of the information recommendation processing method provided by the application. When the heart gesture is scanned in the guide picture 401H, the half screen 402H is popped up to display the interaction result including the instant discount coupon, or the half screen 403H is popped up to display the interaction result including the red envelope, or the half screen 404H is popped up to display the interaction result including the exchange coupon, or the half screen 405H is popped up to display the interaction result including the blessing video, or the interaction result can include the blessing video and the above exchange coupon, the red envelope, and the instant discount coupon at the same time. The interaction result includes but is not limited to the coupon, the instant discount coupon, the red envelope cover, the real object, and the like.

[0165] See Figure 4I ,Figure 4I is an interface schematic diagram of the information recommendation processing method provided by the application embodiment. After receiving a click operation of a user on an interactive result, a card, a direct discount coupon, a red envelope cover, or a real object can be obtained. After receiving a user's obtaining operation on the direct discount coupon in the half screen 401I, the obtaining page 404I is presented. After receiving a user's obtaining operation on the exchange coupon in the half screen 402I, the obtaining page 405I is presented. In the obtaining page 405I, address information is received to complete the obtaining process. After receiving a user's obtaining operation on the red envelope in the half screen 403I, the obtaining page 406I is presented.

[0166] Referring to Figure 4J , Figure 4J is an interface schematic diagram of the information recommendation processing method provided by the application embodiment. When the half screen is loading or fails to load, the toast information of the application embodiment is used for prompting. Figure 4J The prompt information 401J is used to prompt that the network environment is poor, please try again later. The prompt information 402J is used to prompt that the inventory award coupon has been obtained. The prompt information 403J is used to prompt that the award coupon cannot be obtained temporarily.

[0167] Referring to Figure 4K , Figure 4K is an interface schematic diagram of the information recommendation processing method provided by the application embodiment. The guide picture 401K is presented directly through the scanning identification function of the client, and guide text and a background picture are presented in the guide picture. In the guide picture 402K, a heart gesture is identified. When the heart gesture is scanned in the guide picture 403K, a heart-shaped animation, a prompt sound, and vibration appear in the guide picture 403K, and the half screen 404K is popped up to display the interactive result (a welfare type).

[0168] Referring to Figure 4L , Figure 4L is an interface schematic diagram of the information recommendation processing method provided by the application embodiment. When the heart gesture is scanned in the guide picture 401L, a heart-shaped animation, a prompt sound, and vibration appear in the guide picture 401L, and the half screen 402L is popped up to display the interactive result (a welfare type). The half screen can be pulled up to the page 403L. The interactive result includes but is not limited to a card, a direct discount coupon, a red envelope cover, and a real object. After receiving a click operation of a user on the interactive result, the card, the direct discount coupon, the red envelope cover, or the real object can be obtained. When the interactive result is an exchange coupon, the obtaining page 404L is presented, and a trigger obtaining operation is received in the obtaining page 404L. The exchange page 405L is presented to complete the obtaining process.

[0169] In some embodiments, referring to Figure 5 , Figure 5 is a recognition flowchart of the information recommendation processing method provided by the application embodiment. Referring to Figure 7 ,Figure 7 is the identification flowchart of the information recommendation processing method provided by the application embodiment. First, palm detection is performed, then the palm region of interest is obtained after palm detection, and the palm region of interest is presented through the dashed box 701A in Figure 7 . Then, the skeleton positioning processing is performed on the content in the dashed box, the key point skeleton 702A in Figure 7 is obtained through 21 three-dimensional positioning processing, and the gesture classification result is love after the skeleton positioning processing based on the positioning result.

[0170] In some embodiments, referring to Figures 6A-6B , Figures 6A-6B is the model schematic diagram of the information recommendation processing method provided by the application embodiment. A single-point detector is used as a palm detection framework, a BlazeBlock-like basic unit is used to construct a lightweight and efficient backbone network structure, and the basic unit is as shown in Figures 6A-6B . The difference between the two figures is that the step in Figure 6A is 1, and the step in Figure 6B is 2. The multi-scale output of the improved single-point detector is single-head output to reduce the complexity of post-processing (maximum threshold) and calculation. A 21 three-dimensional skeleton regression model is constructed through a backbone network structure similar to the SSD detector, the classification probability of the gesture is output while the skeleton key point positioning is regressed, and the auxiliary model learning is suppressed while the false detection is inhibited. Finally, the normalized feature vectors are obtained by normalizing the three-dimensional skeleton coordinates in each dimension (subtracting the mean value and dividing the variance), and the classification of 16 kinds of gestures (including: "one", "two", "three", "four", "five", "six", "seven", "eight", "first", "good", "bar", "a little bit", "nine", "love", "rock and roll", "love") is realized through a three-layer fully connected (FC+ReLU) multilayer perceptron.

[0171] In some embodiments, the implementation process of the information recommendation processing method provided by the embodiments of the present application is as follows: the advertising end issues advertising configuration information and a detection task to the client, the client sends the collected image and the advertising configuration information to the computing server, the computing server returns the recognition result to the client, the client displays the recognition result and jumps to present the interactive result, the interactive result includes electronic vouchers and the like, and the advertising configuration information includes advertising material information, display time and a set jump gesture, that is, the client executes the detection task within the display time, and the display time restricts the start time of the detection task and the end time of the detection task. The detection task is executed by the client, and when the information stream is presented on the client: after receiving a trigger operation in a legal time period (display time period), the camera is called to start gesture detection; in the process of collecting gestures, valid images are selected through a “still frame” and a “clear frame” strategy, the “still frame” strategy restricts that the frame difference between adjacent frames is less than a frame difference threshold, and the “clear frame” strategy restricts that the variance of the first-order gradient of the frame is less than a clarity threshold, that is, the frame that is clear and still is determined as a valid image and is sent to the background computing server; the computing server identifies the corresponding gesture type through a gesture detection algorithm and returns a jump flag; after the client receives the jump flag, the interactive logic is triggered to present the interactive result, and if the corresponding gesture is not recognized, the detection is restarted.

[0172] Through the information recommendation processing method provided by the embodiments of the present application, the attractiveness of the information stream advertising material can be improved, the interactive interest of the advertisement can be increased, and the advertisement click conversion efficiency can be improved.

[0173] The following continues to illustrate an exemplary structure of the implementation of the information recommendation processing apparatus 455 provided by the embodiments of the present application as a software module. In some embodiments, as shown in FIG. 4, the software module stored in the information recommendation processing apparatus 455 of the memory 450 can include: a display module 4551 configured to display an information stream page; a trigger module 4552 configured to display a collected image in response to an image recognition trigger operation received in the information stream page; and an interactive module 4553 configured to present an interactive result related to a to-be-recommended item when the collected image includes an interactive object and the interactive object is associated with the to-be-recommended item in the information stream page. Figure 2

[0174] In the above scheme, the display module 4551 is further configured to display a first-level information stream page; and the first-level information stream page is a default information stream page displayed when the client is started, and the information stream in the first-level information stream page includes at least one of the following: at least one historical session in which a login account participates; a subscription message of the login account; and a notification message of the login account.

[0175] ​In the above solutions, the display module 4551 is further configured to display an image recognition portal in the first-level information stream page; and determine, as the image recognition trigger operation, a trigger operation received for the image recognition portal.

[0176] In the above solutions, the display module 4551 is further configured to display a second-level information stream page; and the information stream of the second-level information stream page comprises social dynamics of the login account.

[0177] In the above solutions, the display module 4551 is further configured to insert and display at least one recommended information in the information stream of the second-level information stream page; and determine, as the image recognition trigger operation, a trigger operation received for the recommended information.

[0178] In the above solutions, the display module 4551 is further configured to insert and display at least one recommended information in the information stream of the second-level information stream page; and in response to a trigger operation for the recommended information, present a detail page of the recommended information, display an image recognition portal in the detail page, and determine, as the image recognition trigger operation, a trigger operation received for the image recognition portal.

[0179] In the above solutions, the apparatus 455 further includes a sharing module 4554 configured to: in response to a sharing operation for the interaction result, send a sharing message corresponding to the interaction object; and the sharing message is configured to jump to the image recognition portal when triggered.

[0180] In the above solutions, when the image recognition portal is displayed, the display module 4551 is further configured to display first prompt information to prompt the interaction object to be collected; and the first prompt information comprises at least one of the following: introduction information of the interaction object, and a diagram of the interaction object.

[0181] In the above solutions, the display module 4551 is further configured to: when the collected image does not include the interaction object associated with the item to be recommended in the information stream page, present second prompt information to prompt to continue image collection; and when the collected image includes the interaction object associated with the item to be recommended, present a special effect corresponding to the interaction object; and the special effect comprises at least one of the following: an animation, a prompt sound, and a vibration feedback.

[0182] In the above solutions, the interaction module 4553 is further configured to: identify a state of the interaction object, and present an interaction result associated with the state of the interaction object; and the interaction object comprises at least one of the following: a limb, a hand, and a face; the state of the interaction object comprises at least one of the following: a limb action, a hand gesture, and a facial expression; and the interaction result comprises at least one of the following: an electronic red packet, an electronic coupon, and an electronic discount coupon.

[0183] In the above solution, the interaction module 4553 is further configured to: determine an effective image for target recognition from the collected image; invoke an object recognition model to perform target recognition on the effective image to determine a type of the object included in the effective image; and invoke a state type recognition model to perform key point positioning processing on the object included in the effective image to obtain a state of the object.

[0184] In the above solution, the interaction module 4553 is further configured to: when the collected image is a static image collected by photographing, determine the static image as the effective image for target recognition; and when the collected image is a dynamic video frame collected by video shooting, determine a video frame that meets at least one of the following conditions from a plurality of video frames as the effective image for target recognition: a mean value of frame differences between the video frame and adjacent video frames is less than a still frame threshold; and a variance of a first-order gradient of the video frame is greater than a definition threshold.

[0185] In the above solution, the interaction module 4553 is further configured to: before determining the effective image for target recognition from the collected image, obtain a recommended effective period; and when a collection time of the collected image is within the recommended effective period, determine that the operation of determining the effective image for target recognition from the collected image will be performed.

[0186] In the above solution, the interaction module 4553 is further configured to: determine a difference between the recognized interaction object and the interaction object illustration from the collected image, determine a score of the collected image according to the difference, and present an interaction result associated with the score; and the difference includes at least one of the following: a position difference between the recognized interaction object and the interaction object illustration; a state difference between the recognized interaction object and the interaction object illustration; and an image quality difference between the recognized interaction object and the interaction object illustration.

[0187] In the above solution, the interaction module 4553 is further configured to: determine a score of each collected image based on a difference between the recognized interaction object and the interaction object illustration from each collected image; determine a number of images whose scores exceed a score threshold, and present an interaction result associated with the number.

[0188] The embodiment of the present application provides a computer program product or a computer program, which comprises computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the information recommendation processing method in the embodiment of the present application.

[0189] The embodiment of the present application provides a computer readable storage medium storing executable instructions, wherein the executable instructions are stored, and when the executable instructions are executed by a processor, the processor executes the method provided by the embodiment of the present application, for example, as shown in the following. Figures 3A-3C The information recommendation processing method is shown.

[0190] In some embodiments, the computer readable storage medium can be FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disc, or CD-ROM, etc. memory; and can also be various devices including one or any combination of the above memories.

[0191] In some embodiments, the executable instructions can be in the form of programs, software, software modules, scripts or codes, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and can be deployed in any form, including being deployed as independent programs or being deployed as modules, components, subroutines or other units suitable for use in a computing environment.

[0192] As an example, the executable instructions can but not necessarily correspond to files in a file system, can be stored in part of a file storing other programs or data, for example, stored in one or more scripts in a Hyper Text Markup Language (HTML) document, stored in a single file dedicated to the program in question, or stored in multiple cooperative files (for example, files storing one or more modules, subroutines or code portions).

[0193] As an example, the executable instructions can be deployed to execute on one computing device, or on multiple computing devices located at one site, or on multiple computing devices distributed at multiple sites and interconnected through a communication network.

[0194] In summary, the embodiment of the present application presents the interaction result related to the to-be-recommended item in the way of human-computer interaction, which can not only enrich the presentation mode of the to-be-recommended item to improve the recommendation efficiency of the to-be-recommended item, but also provide diversified user interaction experience, enrich the interactive fun when reading the information related to the to-be-recommended item, and thus increase the interestingness of reading the information related to the to-be-recommended item.

[0195] The above is only an embodiment of the present application, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement and improvement made within the spirit and scope of the present application are included in the protection scope of the present application.

Claims

1. An information recommendation processing method characterized by comprising: The method comprises: displaying an information stream page, wherein the information stream page comprises a first-level information stream page and a second-level information stream page, the first-level information stream page is displayed by default when a social media client is started, and comprises a social information stream in the social media client having a social relationship with a login account; the second-level information stream page comprises social dynamics of the login account and accounts having a social relationship with the login account; in response to an image recognition trigger operation received in the information stream page, displaying a captured image; when the captured image does not comprise an interactive object associated with an item to be recommended in the information stream page, presenting second prompt information to prompt to continue image capturing; when the captured image comprises an interactive object, and the interactive object is associated with an item to be recommended in the information stream page, presenting a special effect corresponding to the interactive object, and presenting an interactive result associated with a state of the interactive object, wherein the state of the interactive object comprises at least one of the following: a limb action, a gesture, and a facial expression.

2. The method of claim 1, wherein: the information stream in the first-level information stream page comprises at least one of the following: at least one historical session in which the login account participates; a subscription message of the login account; and a notification message of the login account.

3. The method of claim 1, wherein, The method further comprises: displaying an image recognition entry in the first-level information stream page; determining that a trigger operation received for the image recognition entry is the image recognition trigger operation.

4. The method of claim 1, wherein, The method further comprises: inserting and displaying at least one recommended information in an information stream of the second-level information stream page; determining that a trigger operation received for the recommended information is the image recognition trigger operation.

5. The method of claim 1, wherein, The method of displaying an information stream page comprises: inserting and displaying at least one recommended information in an information stream of the second-level information stream page; in response to a trigger operation for the recommended information, presenting a detail page of the recommended information, and displaying an image recognition entry in the detail page; determining that a trigger operation received for the image recognition entry is the image recognition trigger operation.

6. The method according to claim 3 or 5, characterized in that, When the image recognition entry is displayed, the method further comprises: displaying first prompt information to prompt an interactive object to be captured; wherein the first prompt information comprises at least one of the following: introduction information of the interactive object, and a diagram of the interactive object.

7. The method of claim 1, wherein, The special effect comprises at least one of the following: an animation, a prompt sound, and a vibration feedback.

8. The method of claim 1, wherein, Before the interactive result associated with the state of the interactive object is presented, the method further comprises: identifying the state of the interactive object; wherein the interactive object comprises at least one of the following: a limb, a hand, and a face; the interactive result comprises at least one of the following: an electronic red packet, an electronic exchange coupon, and an electronic discount coupon.

9. The method of claim 1, wherein, The method further comprises: determining an effective image for target recognition from the captured image; calling an object recognition model to perform target recognition on the effective image to determine a type of an object included in the effective image; The state type recognition model is called to perform key point positioning on an object included in the effective image, to obtain a state of the object.

10. The method of claim 9, wherein, The effective image for target recognition is determined from the collected image, including: When the collected image is a static image collected by photographing, the static image is determined as the effective image for target recognition; When the collected image is a dynamic video frame collected by video shooting, a video frame that meets at least one of the following conditions among a plurality of video frames is determined as the effective image for target recognition: The average of frame differences between the video frame and adjacent video frames is less than a static frame threshold; The variance of the first-order gradient of the video frame is greater than a definition threshold.

11. The method of claim 10, wherein, Before determining the effective image for target recognition from the collected image, the method further includes: Obtaining a recommended effective period; When the collection time of the collected image is within the recommended effective period, it is determined that the operation of determining the effective image for target recognition from the collected image will be performed.

12. An information recommendation processing apparatus characterized by comprising: Including: A display module is configured to display an information stream page; wherein the information stream page includes a first-level information stream page and a second-level information stream page, the first-level information stream page is displayed by default when a social media client is started, and includes social information streams in the social media client that have a social relationship with a login account; the second-level information stream page includes social dynamics of the login account and accounts that have a social relationship with the login account; A trigger module is configured to display a collected image in response to an image recognition trigger operation received in the information stream page; An interaction module is configured to present a second prompt information to prompt to continue image collection when the collected image does not include an interaction object associated with an item to be recommended in the information stream page, and present a special effect corresponding to the interaction object and an interaction result associated with a state of the interaction object when the collected image includes the interaction object and the interaction object is associated with the item to be recommended in the information stream page, wherein the state of the interaction object includes at least one of the following: a body action, a gesture, and a facial expression.

13. The apparatus of claim 12, wherein, The display module is further configured to: Display an image recognition entry in the first-level information stream page; Determine a trigger received for the image recognition entry as the image recognition trigger operation.

14. The apparatus of claim 12, wherein, The display module is further configured to: Insert and display at least one recommendation information in an information stream of the second-level information stream page; Determine a trigger received for the recommendation information as the image recognition trigger operation.

15. An electronic device, comprising: Including: A memory is configured to store executable instructions; A processor is configured to execute the executable instructions stored in the memory to implement the information recommendation processing method in any one of claims 1 to 11.

16. A computer-readable storage medium, characterized in that, Executable instructions are stored, and when executed by a processor, the information recommendation processing method in any one of claims 1 to 11 is implemented.

17. A computer program product comprising computer programs or instructions, characterized in that, The computer program or instructions, when executed by a processor, implement the information recommendation processing method of any one of claims 1 to 11.

Images