Content searching method and device

Abstract

The embodiment of the invention provides a content search method and device.The content search method comprises the steps that a search page for a target image is displayed, and the search page comprises at least one image recognition result, a summary content display area and a search result display area; the image recognition result is obtained by performing image recognition on the target image; and in response to a selection operation on a target image recognition result in the at least one image recognition result, displaying summary information of the target image recognition result in the summary content display area, and displaying a search result of the target image recognition result in the search result display area. At least one image recognition result is displayed in the search page for selection, when the target image recognition result is selected, the summary information of the target image recognition result and the search result of the target image recognition result are displayed, the displayed content is rich, and related information obtained by searching the target image can be quickly obtained through the summary information, so that the search efficiency is improved. The information acquisition efficiency is improved, and the user experience is ensured.

Description

Technical Field

[0001] The embodiments of this specification relate to the field of computer technology, and more particularly to a content search method and device. Background Art

[0002] With the continuous development of computer and Internet technology, more and more content application platforms are loved by the majority of users. Content application platforms are community platforms that provide users with information sharing, dissemination and acquisition. Users can view content posted by other users on the content application platforms, such as notes, topics, live broadcasts, etc.

[0003] In the prior art, when users search in a content application platform, the searched-for content is often displayed. Users can swipe up and down to view the searched-for content one by one. The displayed content is relatively simple, which is not conducive to users quickly obtaining search results. The information acquisition efficiency is low, which affects the user experience. Summary of the Invention

[0004] In view of this, embodiments of this specification provide a content search method. One or more embodiments of this specification also relate to another content search method, a content search method apparatus, another content search apparatus, a computing device, a computer-readable storage medium, and a computer program product to address technical deficiencies in the prior art.

[0005] According to a first aspect of an embodiment of this specification, a content search method is provided, which is applied to a client and includes:

[0006] Displaying a search page for a target image, the search page including at least one image recognition result, a summary content display area, and a search result display area; the image recognition result is obtained by performing image recognition on the target image;

[0007] In response to a selection operation of a target image recognition result among the at least one image recognition result, summary information of the target image recognition result is displayed in the summary content display area, and search results of the target image recognition result are displayed in the search result display area.

[0008] According to a second aspect of an embodiment of this specification, a content search method is provided, which is applied to a server and includes:

[0009] In response to a search event for a target image, performing image recognition on the target image to obtain at least one image recognition result, generating a search page for the target image, and returning the search page to the client, the search page including the at least one image recognition result, a summary content display area, and a search result display area;

[0010] Determining summary information corresponding to a target image recognition result and search results corresponding to the target image recognition result;

[0011] The summary information of the target image recognition result and the search results of the target image recognition result are returned to the client, so that the client displays the summary information of the target image recognition result in the summary content display area and displays the search results of the target image recognition result in the search result display area.

[0012] According to a third aspect of the embodiments of this specification, a content search device is provided, which is applied to a client and includes:

[0013] A first display module is configured to display a search page for a target image, the search page including at least one image recognition result, a summary content display area, and a search result display area; the image recognition result is obtained by performing image recognition on the target image;

[0014] The second display module is configured to display summary information of the target image recognition result in the summary content display area and display search results of the target image recognition result in the search result display area in response to a selection operation of the target image recognition result in the at least one image recognition result.

[0015] According to a fourth aspect of the embodiments of this specification, a content search device is provided, which is applied to a server and includes:

[0016] a first returning module configured to, in response to a search event for a target image, perform image recognition on the target image to obtain at least one image recognition result, generate a search page for the target image, and return the search page to the client, the search page including the at least one image recognition result, a summary content display area, and a search result display area;

[0017] a determination module configured to determine summary information corresponding to a target image recognition result and a search result corresponding to the target image recognition result;

[0018] The second returning module is configured to return the summary information of the target image recognition result and the search results of the target image recognition result to the client, so that the client displays the summary information of the target image recognition result in the summary content display area and displays the search results of the target image recognition result in the search result display area.

[0019] According to a fifth aspect of the embodiments of this specification, there is provided a computing device, including:

[0020] memory and processor;

[0021] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the above-mentioned content search method are implemented.

[0022] According to a sixth aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of the above-mentioned content search method are implemented.

[0023] According to a seventh aspect of the embodiments of this specification, a computer program product is provided, comprising a computer program / instruction, which implements the steps of the above-mentioned content search method when executed by a processor.

[0024] One embodiment of this specification provides a content search method. When searching for a target image, a search page for the target image can be displayed. The search page includes at least one image recognition result obtained by performing image recognition on the target image, as well as a summary content display area and a search result display area. The user can select the target image recognition result, and the summary information of the target image recognition result is displayed in the summary content display area, and the search results of the target image recognition result are displayed in the search result display area. In this way, at least one image recognition result obtained by recognition can be displayed on the search page for selection. When the target image recognition result is selected, the summary information of the target image recognition result and the search results of the target image recognition result can be displayed. The content displayed on the search page is rich, and the summary information can be used to quickly obtain relevant information obtained by searching for the target image, thereby improving information acquisition efficiency and ensuring user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a flowchart of a content search method provided by one embodiment of this specification;

[0026] Figure 2a This is a schematic diagram of the first method of displaying a search entry on a content details page provided by an embodiment of this specification;

[0027] Figure 2b This is a schematic diagram of a second method of displaying a search entry on a content details page provided by an embodiment of this specification;

[0028] Figure 2c This is a schematic diagram of a third method of displaying a search entry on a content details page provided by an embodiment of this specification;

[0029] Figure 2d is a schematic diagram of a first target image search page provided by an embodiment of this specification;

[0030] Figure 2eis a schematic diagram of a second target image search page provided by an embodiment of this specification;

[0031] Figure 2f This is a schematic diagram of interactive links related to search content in the first type of summary information provided by an embodiment of this specification;

[0032] Figure 2g This is a schematic diagram of interactive links related to search content in the second type of summary information provided by an embodiment of this specification;

[0033] Figure 2h This is a schematic diagram of interactive links related to search content in the third type of summary information provided by an embodiment of this specification;

[0034] Figure 2i This is a schematic diagram of interactive links related to search content in the fourth type of summary information provided by an embodiment of this specification;

[0035] Figure 3 is a flowchart of another content search method provided by an embodiment of this specification;

[0036] Figure 4a This is a schematic diagram of the processing process of a content search method provided by an embodiment of this specification in a screenshot scenario;

[0037] Figure 4b This is a schematic diagram of an interactive interface of a content search method provided by an embodiment of this specification in a screenshot scenario;

[0038] Figure 5a This is a schematic diagram of the processing process of a content search method provided by an embodiment of this specification in a comment scenario;

[0039] Figure 5b This is a schematic diagram of an interactive interface of a content search method provided by an embodiment of this specification in a screenshot scenario;

[0040] Figure 6 This is a flow chart of a content search method provided by an embodiment of this specification;

[0041] Figure 7 This is a schematic diagram of the structure of a content search device provided by an embodiment of this specification;

[0042] Figure 8 This is a schematic diagram of the structure of another content search device provided by an embodiment of this specification;

[0043] Figure 9 This is a structural block diagram of a computing device provided by one embodiment of this specification. DETAILED DESCRIPTION

[0044] The following description sets forth many specific details to facilitate a thorough understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the scope of this specification. Therefore, this specification is not limited to the specific implementations disclosed below.

[0045] The terms used in one or more embodiments of this specification are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of this specification. The singular forms "a," "an," and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0046] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0047] In addition, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in one or more embodiments of this specification are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0048] First, the terms involved in one or more embodiments of this specification are explained.

[0049] In response to: Indicates a state where a corresponding event occurs or a condition is satisfied. The timing of the subsequent action executed in response to the event or condition is not necessarily strongly correlated with the time when the event occurs or the condition is satisfied. For example, in some cases, the subsequent action may be executed immediately when the event occurs or the condition is satisfied; in other cases, the subsequent action may be executed some time after the event occurs or the condition is satisfied.

[0050] Content application platform: An application platform that integrates multiple contents. The content application platform provides users with rich and diverse multimedia content, including but not limited to notes, topics, live broadcasts, on-demand videos, audio, graphic information, social interactions, etc.

[0051] Published content: refers to content published on community applications. Published content is a form of content provided by the content application platform that allows users to publish created content for viewing by other users. For example, published content can be graphic information (such as notes), live broadcasts, comments, statements or discussions published by users on the content application platform.

[0052] Trigger Action: refers to clicks, swipes, or other interactive actions on relevant content, controls, or components within display interfaces such as content application platforms, published content, and live streaming interfaces. This triggers a corresponding event and displays the interface or information corresponding to the trigger action. For example, clicking on a target interactive information block in the interactive area of ​​a search results page (i.e., a trigger action on the target interactive information block) will display the interactive details page for the target interactive information block; clicking on multimedia content in the interactive area of ​​a search results page (i.e., a trigger action on multimedia content) will display the corresponding associated page.

[0053] Control: In a computer user interface, a "control" is any element with which a user can interact directly. These elements allow users to enter data, select options, or initiate actions. Controls can be graphical (such as buttons, text boxes, checkboxes, etc.) or more abstract (such as voice commands). Buttons typically perform an action or open a new screen; text boxes allow users to enter or edit text; checkboxes represent Boolean values ​​(yes / no) that users can select or deselect; drop-down lists offer a list of options for users to choose from, typically displaying only one selected value; list boxes display multiple options, allowing users to select one or more items; labels display static text, typically describing the function of other controls, but can also function as buttons; scrollbars enable users to navigate large data sets, such as browsing long documents or lists; images display static images and can sometimes function as buttons; and combo boxes combine the functionality of text boxes and drop-down lists, allowing users to enter text and select from a list. These controls are usually provided with standard styles by the operating system or development toolkit, and developers can customize their appearance and behavior as needed. Different operating systems and programming languages ​​(such as Java, C#, Python, etc.) have their own set of controls and provide corresponding APIs to create and manage these controls.

[0054] Multimedia content refers to content that includes multiple media formats, including but not limited to text, images, audio, and video. Multimedia technology integrates these different media types to provide a richer and more interactive user experience. Search for target content. The multimedia content displayed in the interactive area of ​​the search results page can be promotional materials configured for the target account.

[0055] In this specification, a content search method is provided. This specification also involves another content search method, a content search device, another content search device, a computing device, a computer-readable storage medium, and a computer program product, which are described in detail one by one in the following embodiments.

[0056] See also Figure 1 , Figure 1 A flow chart of a content search method provided according to an embodiment of the present specification is shown, which is applied to a client and specifically includes the following steps 102 to 104 .

[0057] Step 102: Display a search page for the target image, the search page including at least one image recognition result, a summary content display area, and a search result display area; the image recognition result is obtained by performing image recognition on the target image.

[0058] Specifically, the target image is an image to be searched. When a user views published content in a content application platform, he or she may be interested in an image and want to search for the image. The image is the target image.

[0059] In actual implementation, the search page refers to a page generated by searching with the target image as the search target and then integrating the relevant results obtained from the search. The search page includes at least one image recognition result, a summary content display area and a search result display area.

[0060] The image recognition result is the result of image recognition of the target image, and is used to indicate information such as the name, category, and location of the object in the target image. For example, the image recognition result may be the recognition result of the item, place, brand, etc. included in the target image. Specifically, the user triggers a search based on the target image on the client, and the client can send a search request for the target image to the server. The server analyzes and recognizes the target image through image recognition tools (image analysis models, specific recognition algorithms, etc.), such as feature extraction and pattern matching, to identify various elements, objects, scenes, and other information in the target image, obtain at least one image recognition result included in the target image, and feed it back to the client.

[0061] It should be noted that the summary content display area is a specific area used to display summary information of relevant search results, so that users can quickly understand the core content of the search results without having to view all the detailed search results one by one.

[0062] The search results display area is used to present the specific results obtained from the search, such as the searched graphic information (such as notes), live broadcasts, comments, statements or discussions and other published content. The search results display area can provide users with an intuitive and comprehensive search results display. Users can quickly browse to matching search results of various types through this area. The search results display area can also support users to sort and filter search results to meet the user's personalized search needs. Specifically, the search results displayed in the search results display area can be obtained based on image recognition results. For example, if the image recognition results include product A, then a search can be performed based on product A, and the corresponding search results are various published content related to product A.

[0063] In an optional implementation of this embodiment, the content application platform may monitor the user's search intention for the target image, that is, before displaying the search page for the target image, further comprising:

[0064] In response to an interactive event on a content details page of a target published content, a search entry is displayed on the content details page; the content details page includes a target image;

[0065] Displays the search page for the target image, including:

[0066] In response to a triggering operation of the search portal, a search page for the target image is displayed.

[0067] It should be noted that users can view the content shared by others on the content application platform. If the user is interested in the images in the content, such as the objects shared in the images, he or she can leave a message in the comment area of ​​the content to ask the publisher, or search for keywords outside the platform or take screenshots to search on other platforms. In the process of viewing the content, it is very inconvenient for users to search for information, the information search efficiency is low, and the operation is cumbersome, which affects the user experience.

[0068] In actual implementation, the interactive events of the content details page of the target published content can be monitored. The content details page includes a target image, so as to display a search entrance in the content details page. Based on the search entrance, the target image in the content details page can be searched for with one click to display the corresponding search page.

[0069] Interaction events on the content details page are events triggered when a user views the content details page of a target publication. These interaction events can reflect the user's search intent. The search portal is a search control that allows users to directly search for target images on the content details page.

[0070] In the embodiments of this specification, when a user views the content details page of the target published content, the user's search intention for the target image in the content details page can be monitored, and a corresponding search entry can be provided to implement one-click search, thereby ensuring that the user can quickly and conveniently obtain relevant search information, thereby improving the user's certainty in obtaining information, facilitating user operations, improving search efficiency, ensuring user experience, and also serving as another content distribution channel for content application platforms.

[0071] It should be noted that in addition to monitoring user interaction events in the content details page and implementing one-click search in the content details page, it is also possible to monitor user search intentions in other scenarios and provide a search entry.

[0072] For example, in a live broadcast room scenario, user interaction events in the live broadcast room can be monitored, such as staying in the live broadcast room for more than a set time, posting comments in the live broadcast room, giving gifts in the live broadcast room, taking screenshots in the live broadcast room, etc. In this case, a search portal can be displayed on the live broadcast screen of the live broadcast room. In response to the triggering operation of the search portal, a search page for the live broadcast screen in the live broadcast room (i.e., the target image) is displayed. Alternatively, in an e-commerce scenario, interaction events on the e-commerce homepage can be monitored, such as browsing a certain product for more than a set time, clicking / collecting / sharing a certain product. In this case, a search portal for a certain product can be displayed on the e-commerce homepage. In response to the triggering operation of the search portal, a search page for the product (the product image is the target image) is directly displayed.

[0073] In an optional implementation of this embodiment, the interactive event on the content details page includes at least one of a view event, a comment event, and a screenshot event; and in response to the interactive event on the content details page of the target published content, displaying a search entry on the content details page includes:

[0074] In response to the viewing event, displaying a search entry at the image and / or text of the content details page;

[0075] In response to a comment event, a search entry is displayed on the comment page of the content details page;

[0076] In response to the screenshot event, a search entry is displayed on the screenshot interface of the content details page.

[0077] In actual implementation, the interactive events of the content details page include at least one of viewing events, comment events, and screenshot events. A viewing event refers to a user browsing the content details page for a set time or performing a set interactive operation, such as a user browsing the content details page for 30 seconds, a user repeatedly sliding the content details page up and down, a user sharing / collecting the target published content, etc., then the viewing event is monitored. Detecting the viewing event indicates that the user is more interested in the content details page and may want to further search for more relevant information. A comment event refers to an event used to post comments in the comment area under the content details page. Users can generally ask relevant questions by posting comments. Therefore, when a comment event is detected, it indicates that the user may want to search for relevant information on the content details page. A screenshot event refers to an event in which a user captures the content details page while viewing the content details page. Users can also search for information through screenshots. Therefore, when a screenshot event is detected, it indicates that the user may want to search for relevant information on the content details page.

[0078] It should be noted that the search entry's location can vary depending on the interaction event. Specifically, when a view event is detected, the search entry can appear within the image, within the text, or both. When a comment event is detected, the search entry will appear within the comment section of the content detail page. When a screenshot event is detected, the search entry will appear on the screenshot interface of the content detail page.

[0079] Specifically, the display format of the search entry can also be adjusted according to different interactive events. For example, when a view event is detected, the search entry can be a prominent button or icon that is easy for users to click quickly; when a comment event is detected, the search entry can be a link below or next to the comment area to guide users to search; for screenshot events, the search entry can be a prompt on the screenshot interface, prompting users to search by clicking on the prompt.

[0080] For example, Figure 2a This is a schematic diagram of the first method of displaying a search entry on a content details page provided by an embodiment of this specification, such as Figure 2a As shown, the viewing event of the content details page is monitored, and the search entry "Click to search for the same item" is displayed at the target image of the content details page. Figure 2b This is a schematic diagram of a second method of displaying a search entry on a content details page provided by an embodiment of this specification, such as Figure 2b As shown, the comment event of the content details page is monitored, and the search entry "Try to search for the same item" is displayed on the comment page of the content details page. Figure 2c This is a schematic diagram of a third method of displaying a search entry on a content details page provided by an embodiment of this specification, such as Figure 2c As shown, a screenshot event of the content details page is detected, and a search entry "Search More" is displayed on the screenshot interface of the content details page.

[0081] In the embodiments of this specification, the display position of the search entrance on the content details page will change according to different interactive events. Based on the different interactive behaviors of users, the display position of the search entrance will be intelligently adjusted. When users have different interactive behaviors, it is convenient for users to use the search function at any time, providing users with a more personalized search experience, thereby better meeting users' search needs.

[0082] In an optional implementation of this embodiment, in response to a comment event, a search entry is displayed on the comment page of the content details page, including any of the following methods:

[0083] In response to publishing a target review content including a target image, displaying a search portal on a review page of the content details page;

[0084] In response to the target review content being published, performing intent recognition on the target review content, and when the result of the intent recognition matches the target image, displaying a search entry on the review page of the content details page;

[0085] In response to the target review content being published, extracting the review keywords of the target review content, and when the review keywords are set keywords corresponding to the target image, displaying a search entry on the review page of the content details page;

[0086] In response to the target review content being published, the sentiment type of the target review content is identified, and when the sentiment type is a positive type, a search entry is displayed on the review page of the content details page.

[0087] In an optional implementation, in response to the publication of target comment content including a target image, a search portal can be displayed on the comment page of the content details page. That is, when a user publishes target comment content with a target image in the content details page of the target published content, such as evaluation, consultation and other comment content of a certain photo or picture, indicating that the user wants to obtain information related to the target image, a search portal can be displayed on the comment page corresponding to the content details page to facilitate the user to implement a search query for the target image.

[0088] In another optional implementation, in response to the target comment content being published, the target comment content is subjected to intent recognition, and when the result of the intent recognition matches the target image, a search entry is displayed on the comment page of the content details page. It should be noted that when a user publishes a target comment content, the published target comment content can be analyzed to identify the intention behind the comment content. For example, the result of intent recognition can be a simple expression of opinion, inquiry, search, or recommendation, etc. If the result of intent recognition is an inquiry, search, or recommendation related to the target image, it means that the user's intention in publishing the target comment content is to obtain information related to the target image. At this time, it can be determined that the result of intent recognition matches the target image, and a search entry is displayed on the comment page of the content details page. The search entry is provided to the user at the appropriate time so that the user can further search for information related to the target image, making it easier for the user to obtain more content.

[0089] In another optional implementation, in response to publishing the target comment content, the comment keywords of the target comment content are extracted, and when the comment keywords are the set keywords corresponding to the target image, a search entry is displayed on the comment page of the content details page. The set keywords corresponding to the target image can be keywords that can indicate the user's search intention for the target image, such as the set keywords can be "same style", "link", "want", etc. In actual implementation, the user can post a comment on the comment page while viewing the content details page of the target published content. After the user posts the comment, the client can extract keywords from the published target comment content to determine whether the published target comment content includes the set keywords. If so, it means that the user posted the target comment content in order to obtain information related to the target image, and the search entry can be displayed at the target comment content posted on the comment page of the content details page.

[0090] In another optional implementation, in response to the target comment content being published, the emotional type of the target comment content is identified. When the emotional type is positive, a search entry is displayed on the comment page of the content details page. In actual implementation, after the user publishes the target comment content, the emotional type of the target comment content can be identified. Emotional types are usually positive, negative, neutral, etc. If the emotional type is identified as positive, it means that the user is more interested in the target image in the content details page of the target published content. At this time, a search entry can be displayed on the comment page to provide the user with a further search entry, so that the user can more conveniently find content related to the target image. For example, if the target comment content published by the user is "The skirt in the picture is so beautiful", the emotional type of the target comment content can be identified as positive, and a search entry can be provided to facilitate the user to search for the skirt he likes with one click.

[0091] In actual implementation, the recognition and analysis process of the above-mentioned target comment content can be performed on the client, and the client can also send the comment content posted by the user to the server, which will perform recognition and analysis and feed back the recognition results to the client. Based on the recognition results, the client determines whether to display the search entry.

[0092] Using the above example, Figure 2b As shown, the comment content posted by the user is "Request for link to Figure 2". The client identifies and analyzes the comment, determines that it is an inquiry-type comment content, indicates the user's search intention, determines that the comment content is the target comment content, and displays the search entry "Try to search for the same item" below the target comment content. The user triggers the search entry "Try to search for the same item". At this time, the target image is Figure 2 in the content details page of the target published content, and the search for the same item as Figure 2 can be directly implemented in the content application platform.

[0093] In addition, during actual implementation, the display position and form of the search entry can be dynamically adjusted according to the type of target review content and changes in keywords, making the display of the search entry more flexible and intelligent.

[0094] In the embodiment of this specification, when a user posts a target comment content and it is predicted that the user has a potential search intention, a search entry can be directly displayed at the target comment content, thereby improving the convenience and personalization of the search function and facilitating users to quickly search for content of interest.

[0095] In an optional implementation of this embodiment, a search page for a target image is displayed, the search page including at least one image recognition result, a summary content display area, and a search result display area, including:

[0096] Displaying a search page for a target image, the search page including at least one image recognition result, a summary content display area, and a search result display area, wherein the summary content display area displays overall summary information of the at least one image recognition result and at least one corresponding summary information block;

[0097] The method also includes:

[0098] In response to a triggering operation of a target summary information block in at least one summary information block, a related page corresponding to the target summary information block is displayed.

[0099] In actual implementation, when searching for a target image, a search page for the target image can be displayed. The search page includes at least one image recognition result, a summary content display area, and a search result display area. The overall summary information of at least one image recognition result and at least one corresponding summary information block can be displayed in the summary content display area.

[0100] Among them, the overall summary information of at least one image recognition result is the core information obtained by summarizing the at least one image recognition result, or the overall summary information may also include source content; the summary information block is jumpable content extracted based on the overall summary information, that is, the summary information block can trigger a direct jump to the corresponding associated page, which is the page that the summary information block can directly jump to display.

[0101] Specifically, the image features of each image recognition result can be obtained. The image features are a set of data that describe key information in the image, such as color, texture, shape, etc. The image features of each image recognition result are filled into a preset prompt word template to generate corresponding prompt information, and the prompt information is input into the target generation model. The target generation model will analyze, understand and refine the input prompt information, remove redundant information, extract key content, and summarize it, and finally generate overall summary information of each image recognition result. This overall summary information can concisely summarize the main content of each image recognition result, making it convenient for users to quickly understand the overall situation of the image.

[0102] The target generation model is a model with natural language processing capabilities, such as a large language model, capable of extracting and summarizing core information. This large model can be a pre-trained large model or a large model that has undergone supervised fine-tuning in a search information summarization scenario. This large model can extract and summarize the core information of each image recognition result and output the corresponding overall summary information.

[0103] The preset prompt word template is a prompt text framework designed in advance. It contains some placeholders for filling in specific information. The image features of each image recognition result obtained can be filled in the corresponding positions according to the requirements of the template, thereby generating prompt information containing image feature information.

[0104] For example, Figure 2d This is a schematic diagram of a first target image search page provided by an embodiment of this specification, such as Figure 2d As shown, the image recognition results of the target image displayed in the search page are three restaurants, namely Restaurant A, Restaurant B, and Restaurant C. The image features of Restaurant A, Restaurant B, and Restaurant C can be filled into the preset template to generate prompt information, and then the prompt information is input into the target generation model. The model may output the overall summary information "The author mentioned three restaurants located in XX, namely Restaurant A, Restaurant B, and Restaurant C. The average cost per person in the three restaurants is around 200, which is suitable for group gatherings to try more dishes, and has an atmosphere suitable for dates"; in addition, the overall summary information also includes the generated data source "summarized 12 people's answers and 42 notes generated", and the overall summary information is displayed in the summary content display area.

[0105] It should be noted that at least one summary information block can also be displayed in the summary content display area of ​​the search page. The summary information block can be obtained based on the analysis of the overall summary information, and the key content of the corresponding published content in the overall summary information can be extracted to generate the corresponding summary information block, so that the corresponding related page can be directly jumped based on the summary information block.

[0106] In actual implementation, after generating corresponding overall summary information based on at least one image recognition result, the overall summary information can also be input into the large model for analysis, and the contained keywords can be extracted, and matched based on the keywords in the content application platform. If the corresponding published content can be matched, the corresponding summary information block is generated based on the corresponding keywords.

[0107] Specifically, triggering any summary information block can jump to the corresponding account homepage, search result list and other related pages.

[0108] Using the above example, Figure 2d As shown, by analyzing the overall summary information, it can be determined that the keywords include "Restaurant A," "Restaurant B," and "Restaurant C." Summary information block 1 is generated based on "Restaurant A," summary information block 2 is generated based on "Restaurant B," and summary information block 3 is generated based on "Restaurant C." Assume that a user triggering summary information block 1 can access Restaurant A's account homepage, or can also access a content list of various published content related to Restaurant A.

[0109] In addition, if Figure 2d As shown, the search page also includes a search result display area for displaying search results obtained based on image recognition results.

[0110] In an embodiment of the present specification, the overall summary information of at least one image recognition result and at least one corresponding summary information block can be displayed in the summary content display area of ​​the search page. Through the overall summary information, the user can quickly understand the core content of each image recognition result in the target image, and by triggering the summary information block, the corresponding related page can be quickly jumped, which greatly facilitates the efficiency of the user in viewing information.

[0111] In an optional implementation of this embodiment, the search page further includes a target image; and the method further includes:

[0112] In response to a frame selection operation on a target image, newly added image recognition results are displayed on the search page, wherein the newly added image recognition results are obtained based on recognition of the framed content corresponding to the frame selection operation.

[0113] In actual implementation, the search page may also include a target image, and the target image may include a frame selection control. The frame selection control is a tool that allows users to define a specific area, such as a rectangular frame commonly seen in some image editing or recognition software, so that users can perform frame selection operations on the target image.

[0114] Specifically, the user can drag the selection control in the target image to perform the selection operation and select the content of interest. When the user completes the selection operation, the system will identify the part of the image content circled by the selection control, obtain new image recognition results, and display the new image recognition results on the search page, making it convenient for users to view and further use the new information identified.

[0115] Using the above example, Figure 2d As shown, assuming that the image recognition results initially displayed on the search page include "Restaurant A", "Restaurant B", and "Restaurant C", the search page can also display the target image. Assuming that the user drags the selection control "rectangular box" in the target image, the content within the rectangular box can be identified. Assuming that the newly identified image recognition result is "Dish A", "Dish A" can be newly displayed on the search page.

[0116] In an embodiment of the present specification, the target image can be displayed in the search page of the target image, and the target image can be framed to facilitate users to manually frame the image content. New image recognition results are added to the search page to facilitate users to extract specific information from the target image and search and display it, which facilitates user operation and improves user experience.

[0117] Step 104: In response to a selection operation on a target image recognition result among the at least one image recognition result, display summary information of the target image recognition result in the summary content display area, and display search results of the target image recognition result in the search result display area.

[0118] In actual implementation, at least one image recognition result can be provided in the search page of the target image. The user can select the target image recognition result of his or her interest from at least one image recognition result. At this time, the target image recognition result can be extracted and summarized to obtain summary information of the target image recognition result, and the summary information of the target image recognition result can be displayed in the summary content display area; and, based on the target image recognition result selected by the user, a search can be conducted on the content application platform to obtain corresponding search results. The search results can be published content related to the target image recognition results in the content application platform, such as graphic information (notes), products, videos, topics, etc. The search results of the target image recognition results are displayed in the search result display area, and the user can slide to view each search result.

[0119] It should be noted that when searching on the content application platform based on the target image recognition result selected by the user and obtaining the corresponding search results, the candidate content in the content application platform can be obtained, and information of at least one dimension of each candidate content can be extracted, such as title, text, image, comments and other information. Based on the information of at least one dimension of each candidate content, it is matched with the target image recognition result, and the one with higher relevance is used as the search result corresponding to the target image recognition result.

[0120] Using the above example, Figure 2d As shown in FIG, assuming that the target image recognition result selected by the user is “Restaurant A”, a summary analysis can be performed based on “Restaurant A” to generate corresponding summary information “Restaurant A’s special dishes are XXXX, the environment is XXXX, it is located at XXX location, the transportation is convenient, and it is suitable for family gatherings~”, which is displayed in the summary content display area, as shown in FIG. Figure 2e As shown, Figure 2e This is a schematic diagram of a second target image search page provided by an embodiment of this specification. Figure 2e As shown, based on the search for "Restaurant A" in the content application platform, notes related to "Restaurant A" can be obtained and displayed in the search result display area.

[0121] It should be noted that users can flexibly select the target image recognition they are interested in on the target image search page, so that the summary information of the target image recognition results will be displayed in the summary content display area of ​​the search page, and the search results of the target image recognition results will be displayed in the search result display area, so that users can quickly obtain relevant search information. The operation is flexible and can adapt to the different search needs of users, which greatly enriches the user's search experience. At the same time, it also provides more user interaction points and content display opportunities for the content application platform.

[0122] In an optional implementation of this embodiment, the summary information of the target image recognition result is determined by the following method:

[0123] Generate target prompt information based on the target image recognition result and the preset prompt word template;

[0124] The target prompt information is input into the target generation model to obtain the summary information of the target image recognition result.

[0125] Among them, the preset prompt word template is a prompt text framework designed in advance. It contains some placeholders for filling in specific information. The image features of the target image recognition results obtained can be filled in the corresponding positions according to the requirements of the template, thereby generating target prompt information containing the image feature information of the target image recognition results.

[0126] In actual implementation, the image features of the target image recognition results can be obtained. The image features are a set of data that describe the target image recognition results. The image features of the target image recognition results are filled into the preset prompt word template to generate corresponding target prompt information. The target prompt information is input into the target generation model. The target generation model will analyze, understand and refine the input target prompt information, remove redundant information, extract key content, and summarize it, and finally generate summary information of the target image recognition results.

[0127] In addition, based on the target image recognition results, you can also search in the content application platform to obtain relevant content, such as account information, related published content, etc., and input the relevant content into the target generation model. The target generation model conducts a comprehensive analysis of the target prompt information and the related content to generate corresponding summary information.

[0128] In the embodiments of this specification, the target generation model can be used to obtain summary information of the target image recognition results. The target generation model can be trained and optimized, and under the guidance of the target prompt information, it can accurately understand the target image recognition results and generate refined and accurate summary information to show the core content of the target image recognition results, providing users with accurate, comprehensive and detailed core content, which not only improves the efficiency of information processing, but also improves the accuracy and comprehensiveness of information.

[0129] In an optional implementation of this embodiment, the summary information of the target image recognition result includes at least one related search content; and the method further includes at least one of the following:

[0130] In response to a triggering operation of the first related search content, displaying a related content detail page corresponding to the first related search content, wherein the related content detail page includes at least one of a product detail page, a merchant detail page, a topic detail page, and an activity detail page;

[0131] In response to a triggering operation of the second associated search content, displaying an account homepage corresponding to the second associated search content;

[0132] In response to a triggering operation of the third associated search content, displaying a search content list corresponding to the third associated search content;

[0133] In response to a triggering operation of the fourth related search content, displaying a commodity transaction interface corresponding to the fourth related search content;

[0134] In response to the triggering operation of the fifth associated search content, a search page of the fifth associated search content is displayed.

[0135] Specifically, the summary information of the target image recognition result includes at least one associated search content, which can directly jump to the corresponding page to achieve fast information acquisition. For example, the associated search content can be a keyword in the summary information of the target image recognition result.

[0136] In actual implementation, the associated search content included in the summary of the target image recognition result can be generated based on the user's search intent and / or a set recommendation strategy. The search intent is determined based on at least one dimension of information including historical search behavior, similar user search behavior, interactive events, and / or the target published content. The set recommendation strategy can be a content promotion strategy pre-configured in the content application platform, which recommends content to other users to increase exposure.

[0137] It should be noted that the generation of related search content in the summary information is based on two scenarios: In the first possible implementation, it is determined based on the user's search intent. The user's search intent is determined by information from at least one dimension, including historical search behavior (i.e., past search operations performed by the user on the platform, reflecting the user's long-term interests and preferences), similar user search behavior (search operations of other users with similar characteristics and behaviors to the current user, thereby inferring the content that the current user may be interested in), interactive events (such as user likes, comments, and shares on the platform, which reflect the user's current interests), and target published content (published content associated with the target image, which can provide directly relevant information for the related search). In the first possible implementation, it is determined based on a set recommendation strategy. This is a content promotion strategy pre-configured by the content application platform. The purpose is to recommend content to users and increase content exposure. The platform will push relevant content to users who may be interested based on its own operational goals and algorithmic rules.

[0138] In specific implementations, the associated search content can be a variety of different types of content, which can be associated with different pages to achieve jumps to different links. The first associated search content is a specific object such as a product, merchant, topic, or event. In response to the triggering operation of the first associated search content, the associated content details page corresponding to the first associated search content can be directly jumped to display, such as at least one of the product details page, merchant details page, topic details page, and event details page. Users can directly jump to the corresponding details page by clicking on keywords such as products, merchants, topics, and events.

[0139] The second associated search content is a keyword of a certain account. In response to the triggering operation of the second associated search content, the account homepage corresponding to the second associated search content can be directly jumped to display, that is, clicking the account keyword can directly enter the account homepage.

[0140] The third related search content is a list keyword. In response to the triggering operation of the third related search content, the search content list corresponding to the third related search content is directly jumped to display. The search content list includes multiple related published contents. By clicking the list keyword, you can directly enter the related list.

[0141] The fourth related search content is a product keyword. In response to the triggering operation of the fourth related search content, you can directly enter the product transaction interface corresponding to the fourth related search content, that is, click on the product keyword to realize one-click transaction of the corresponding product.

[0142] The fifth associated search content is a search keyword. In response to the triggering operation of the fifth associated search content, the search page displaying the fifth associated search content is directly jumped to, that is, the search keyword is clicked, and a search is directly performed based on the search keyword to display the corresponding search page.

[0143] For example, Figure 2f This is a schematic diagram of an interactive link of the associated search content in the first summary information provided by an embodiment of this specification, such as Figure 2f As shown, the target image recognition result is restaurant A, and the summary information is "Restaurant A's special dishes are XXXX, the environment is XXXX, it is located at XXX location, with convenient transportation, suitable for family gatherings, and the restaurant sells surrounding tableware~", where "surrounding tableware" is a related search content. Clicking on the related search content can directly display the product details page of "the tableware"; or clicking on the related search content can directly display the transaction page of "the tableware".

[0144] Figure 2g This is a schematic diagram of the interactive links of the associated search content in the second summary information provided by an embodiment of this specification, such as Figure 2g As shown, the target image recognition result is Restaurant A, and the summary information is "Restaurant A's special dishes are XXXX, the environment is XXXX, it is located at XXX location, the transportation is convenient, it is suitable for family gatherings, and the restaurant sells surrounding tableware~", where "Restaurant A" is the related search content. Clicking on the related search content can directly display the account homepage of "Restaurant A".

[0145] Figure 2h This is a schematic diagram of interactive links of associated search content in the third type of summary information provided by an embodiment of this specification, such as Figure 2h As shown, the target image recognition result is Restaurant A, and the summary information is "Restaurant A's featured dishes include Dish A, the environment is XXXX, it is located at XXX location, the transportation is convenient, it is suitable for family gatherings, and the restaurant sells surrounding tableware~", where "Dish A" is the associated search content. Clicking on the associated search content can directly display the search content list of "Dish A", and various notes related to "Dish A" are displayed in the search content list.

[0146] Figure 2i This is a schematic diagram of the interactive links of the associated search content in the fourth summary information provided by an embodiment of this specification, such as Figure 2i As shown, the target image recognition result is restaurant A, and the summary information is "Restaurant A's featured dishes include dish A and dish B, the environment is XXXX, it is located at XXX location, the transportation is convenient, suitable for family gatherings, and the restaurant sells surrounding tableware~", where "dish B" is the associated search content. Clicking on the associated search content can directly display the search page for "dish B", which can include the summary information corresponding to dish B, search results (related notes on dish B), etc.

[0147] In the embodiment of this specification, the summary information of the target image recognition result includes at least one related search content. The related search content provides a variety of different interactive links, which are more in line with the user's actual search needs, facilitate users to quickly obtain the required search information, and also help the platform promote content.

[0148] In an optional implementation of this embodiment, the search page further includes at least one related search question, which is generated based on the target image recognition result, summary information of the target image recognition result, and / or search results of the target image recognition result; and the method further includes at least one of the following:

[0149] In response to a first triggering operation of a target-related search question, displaying a search page of the target-related search question, wherein the target-related search question is a triggered question among the at least one related search question;

[0150] In response to the second triggering operation of the target-related search question, a question-and-answer interaction page is displayed, and the search results of the target-related search question are displayed on the question-and-answer interaction page.

[0151] In actual implementation, related search questions can also be generated based on the target image recognition results, the summary information of the target image recognition results and / or the search results of the target image recognition results, and displayed on the search page so that users can continue to ask questions with one click.

[0152] Specifically, the generated related search questions can be intelligently sorted to prioritize the question options that are most likely to meet user needs, thereby simplifying the question-asking process and improving interaction efficiency.

[0153] It should be noted that the search page of the target image can provide related search questions. For any target-related search question, through the first trigger operation, you can directly search based on the target-related search question and display the corresponding search page; through the second trigger operation, you can display a question-and-answer interaction page. The question-and-answer interaction page is an in-depth question-and-answer page. The search results of the target-related search question can be displayed on the question-and-answer interaction page, and the question-and-answer interaction page can display a question control, through which continuous question and answer can be realized on the question-and-answer interaction page.

[0154] The first trigger operation and the second trigger operation may be different interactive operations of the target-related search question, such as the first trigger operation being a left swipe and the second trigger operation being a right swipe.

[0155] Using the above example, Figure 2d As shown, related search questions such as "Which dish in Restaurant A is delicious?" and "Which restaurant has convenient transportation, Restaurant A, Restaurant B, or Restaurant C?" can be displayed in the summary content display area. Triggering any related search question can achieve one-click search.

[0156] In an embodiment of the present specification, at least one related search question can be displayed on the search page of the target image, which not only enriches the content of the search page, but also allows users to see related search questions intelligently generated based on their current search content without having to manually enter questions. They can simply click to further explore, which greatly saves time and energy; moreover, these related search questions can also be personalized based on the user's search history and preferences, making the search results more in line with user needs, further enhancing user satisfaction, and greatly improving the user experience.

[0157] In an optional implementation of this embodiment, the search page further includes a dialogue interaction entrance, and the method further includes:

[0158] In response to the triggering operation of the dialogue interaction entrance, a question-and-answer interaction page is displayed, and at least one candidate question is displayed on the question-and-answer interaction page, wherein the at least one candidate question is generated based on at least one of the target user portrait, similar user portraits, information on at least one dimension of the target published content, interactive events, target image recognition results, summary information of the target image recognition results, and search results of the target image recognition results.

[0159] In actual implementation, the search page may also include a dialogue interaction portal, which is used to enter the Q&A interactive page for question and answer. The interactive portal can be a control that can jump to the Q&A interactive page after being triggered.

[0160] Specifically, after the user triggers to enter the Q&A interactive page, multiple candidate questions related to the target image can be displayed on the Q&A interactive page, so that the user can select them with one click to complete the Q&A. If there is no question the user wants to ask among the candidate questions, the user can also manually enter the question in the Q&A interactive page to complete the Q&A.

[0161] At least one candidate question is generated based on at least one of the following: a target user profile, similar user profiles, information on at least one dimension of the target published content, interactive events, target image recognition results, summary information of the target image recognition results, and search results of the target image recognition results. The target user profile is a collection of information that characterizes and describes the characteristics, preferences, behaviors, etc. of the target user, and can reflect the characteristics and needs of the target user; the similar user profile is a profile of a user group that has similar characteristics or behaviors to the target user, and can be used as a reference to infer the possible needs of the target user; information on at least one dimension of the target published content, interactive events, target image recognition results, summary information of the target image recognition results, search results of the target image recognition results, and other information can indicate the search process for the target image, so as to quickly generate relevant candidate questions.

[0162] It should be noted that in response to the triggering operation of the dialogue interaction entrance, the question-and-answer interaction page is entered. At least one candidate question can be provided on the question-and-answer interaction page, which facilitates users to ask questions directly based on the intelligently recommended candidate questions. The operation is simple, which further improves the user experience.

[0163] An embodiment of the present specification provides a content search method, in which at least one image recognition result obtained by identification can be displayed on a search page for selection. When a target image recognition result is selected, summary information of the target image recognition result and search results of the target image recognition result can be displayed. The content displayed on the search page is rich, and the relevant information obtained by searching for the target image can be quickly obtained through the summary information, thereby improving information acquisition efficiency and ensuring user experience.

[0164] See also Figure 3 , Figure 3 A flowchart of another content search method provided according to an embodiment of this specification is shown, which is applied to a server and specifically includes the following steps.

[0165] Step 302: In response to a search event for a target image, image recognition is performed on the target image to obtain at least one image recognition result, a search page for the target image is generated, and the search page is returned to the client. The search page includes at least one image recognition result, a summary content display area, and a search result display area.

[0166] Specifically, the target image search event refers to a search event triggered based on the target image, such as an event that triggers a search entry.

[0167] In actual implementation, when image recognition is performed on the target image to obtain at least one image recognition result, an image recognition prompt word can be obtained. The image recognition prompt word is a pre-configured prompt information used to define which dimensional information in the image is to be recognized. The target image and the image recognition prompt word are input into the image recognition model. The image recognition model can extract the corresponding dimensional information (such as scene, location, product, etc.) from the target image according to the instructions of the image recognition prompt word, identify the content therein, and obtain at least one image recognition result.

[0168] It should be noted that when performing image recognition on a target image, a pre-configured image recognition prompt is obtained. This prompt specifies which dimensions of information to focus on when recognizing the image. For example, it can specify whether to recognize the scene (such as a park or a shopping mall), the location (such as a specific latitude and longitude or a landmark), or the product (such as a certain brand of mobile phone or a certain style of clothing).

[0169] By using prompt words and target generation models, image recognition results can be obtained quickly and accurately, making image recognition more targeted and meeting the needs of recognizing specific dimensional information in different scenarios.

[0170] In actual implementation, after obtaining at least one image recognition result, a search page for the target image can be generated. The specific implementation process is the same as above. Figure 1 The embodiments shown are similar, and the embodiments of this specification will not be repeated here.

[0171] Step 304: Determine summary information corresponding to the target image recognition result and search results corresponding to the target image recognition result.

[0172] In actual implementation, if the user selects a target image recognition result from at least one image recognition result through the client, the server can determine summary information corresponding to the target image recognition result and search results corresponding to the target image recognition result.

[0173] In one implementation, when the server performs image recognition on a target image to obtain at least one image recognition result and generates a search page for the target image, it can search for each image recognition result identified separately to obtain search results corresponding to each image recognition result. The server can also generate target prompt information based on the image recognition result and a preset prompt word template, call a target generation model, and obtain summary information of each image recognition result. When the user selects a target image recognition result, the server can query the summary information corresponding to the target image recognition result and the corresponding search results.

[0174] In another implementation method, after the user selects a target image recognition result from at least one image recognition result through the client, the server can also generate target prompt information based on the target image recognition result selected by the user and combined with a preset prompt word template, call the target generation model, and obtain summary information of the target image recognition result; and, it can search based on the target image recognition result selected by the user to obtain search results corresponding to the target image recognition result.

[0175] It should be noted that the above-mentioned target generation model is deployed on the server side, and the server side calls the target generation model to generate corresponding summary information. In another implementation method, the target generation model can also be directly deployed on the client side. The client directly generates target prompt information based on the target image recognition result and the preset prompt word template, and calls the corresponding target generation model with the target prompt information to obtain summary information of the target image recognition result. The embodiments of this specification do not limit this.

[0176] Step 306: Return the summary information of the target image recognition result and the search results of the target image recognition result to the client, so that the client displays the summary information of the target image recognition result in the summary content display area and displays the search results of the target image recognition result in the search results display area.

[0177] In an embodiment of the present specification, the server can perform image recognition on the target image to obtain at least one image recognition result, generate a search page for the target image, and provide at least one image recognition result for selection on the search page. When the target image recognition result is selected, the server can also generate summary information of the target image recognition result based on the target image recognition result, and search based on the target image recognition result to obtain corresponding search results, and feedback rich content to the client, so that the user can quickly obtain relevant information obtained by searching the target image through the summary information, thereby improving information acquisition efficiency and ensuring user experience.

[0178] The above is a schematic solution for a content search method on the server side of this embodiment. It should be noted that the technical solution of this content search method on the server side and the technical solution of the content search method on the client side described above are based on the same concept. For details not described in detail in the technical solution of the content search method on the server side, please refer to the description of the technical solution of the content search method on the client side described above.

[0179] The following combined Figure 4a and attached Figure 4b , taking the application of the content search method provided in this specification in the screenshot scenario as an example, the content search method is further explained. Figure 4aA schematic diagram of the processing process of a content search method provided by an embodiment of this specification in a screenshot scenario is shown. Figure 4b A schematic diagram of an interactive interface of a content search method provided by an embodiment of this specification in a screenshot scenario is shown. Figure 4a and Figure 4b As shown, the following steps are included.

[0180] Step 402: In response to a screenshot operation on a target image in a content details page of a target published content, a screenshot interface is displayed on the content details page, and a search entry is displayed in the screenshot interface.

[0181] like Figure 4b As shown in the figure, when a user is viewing the content details page of a target published content, he takes a screenshot of the target image. At this time, a screenshot event is detected and a screenshot interface is displayed on the content details page. The screenshot interface displays multiple sharing controls, such as "Private message friends", "XX app friends", and "YY app friends". Users can share the screenshot with friends through the sharing controls. In addition, Figure 4b As shown, the screenshot interface may also display a search entry “Search More”.

[0182] Step 404: In response to the triggering operation of the search entry in the screenshot interface, a search page for the target image is displayed, and the search page includes at least one image recognition result, a summary content display area and a search result display area. The image recognition result is obtained by performing image recognition on the target image.

[0183] like Figure 4b As shown, when the user clicks on "Search More", the search page of the target image (the image captured by the user in the content details page of the target published content) is displayed. The search page displays the image recognition results of "top", "pants", and "bag". The search page includes a summary content display area and a search result display area. The summary content display area displays summary information of the image recognition results in the target image, such as "top is from brand A, pants are from brand B, bag is from brand C", as well as summary information blocks of "top", "pants", and "bag". Triggering the summary information block can enter the corresponding related content list page, transaction page, details page, etc.

[0184] Step 406: In response to a selection operation on a target image recognition result among the at least one image recognition result, display summary information of the target image recognition result in the summary content display area, and display search results of the target image recognition result in the search result display area.

[0185] Continuing with the above example, suppose the user selects "top" and top is the target image recognition result. At this time, the summary information of "top" can be displayed in the summary content display area, such as "top is from brand A, with XX style, very suitable for XX scene~", and notes related to "top" can be displayed in the search result display area.

[0186] It should be noted that when a user views the content details page of a target published content on a content application platform, if a screenshot of a target image in the content details page of the target published content is taken, it indicates that the user may be interested in the target image and wants to conduct further information search based on the screenshot. However, entering the search page based on the screenshot or searching for related content on other platforms is not only cumbersome, but the user may not necessarily find the information he wants. Searching for information is very inconvenient, and the information search efficiency and accuracy are low, which greatly affects the user experience.

[0187] Therefore, one embodiment of this specification provides a content search method. In a screenshot scenario, if a user's screenshot operation of a target image in the content details page of the target published content is detected, a search entry can be directly displayed in the screenshot interface. When the user triggers the search entry, a one-click search for the target image can be realized, and a search page for the target image is displayed. The search page displays at least one image recognition result, summary information, search results, etc. of the target image, and supports the user to flexibly select the target image recognition result of interest from at least one recognized image recognition result, and display the summary information and search results of the target image recognition result. By taking a screenshot of the target image, a search entry can be provided to realize a one-click search for the target image. There is no need to enter the search page based on the screenshot or search for related content on other platforms. This greatly simplifies the user's screenshot search operation, allowing the user to quickly and accurately obtain the information they want to search for, ensuring user experience.

[0188] The above is a schematic solution of a content search method in the screenshot scenario of this embodiment. It should be noted that the technical solution of the content search method in the screenshot scenario is the same as the above Figure 1-Figure 3 The technical solutions of the content search method shown in the figure belong to the same concept. For details not described in detail in the technical solution of the content search method in the screenshot scenario, please refer to the above Figure 1-Figure 3 Description of the technical solution of the content search method shown.

[0189] The following combined Figure 5a and attached Figure 5b , taking the application of the content search method provided in this specification in the comment scenario as an example, the content search method is further explained. Figure 5a A schematic diagram of the processing process of a content search method provided by an embodiment of this specification in a comment scenario is shown. Figure 5b A schematic diagram of an interactive interface of a content search method provided by an embodiment of this specification in a screenshot scenario is shown. Figure 5a and Figure 5b As shown, the following steps are included.

[0190] Step 502: In response to the publishing operation of the target comment content in the content details page of the target published content, if the target comment content includes a target image, a search entry is displayed on the comment page of the content details page; and / or, if the result of the intent recognition of the target comment content matches the target image, a search entry is displayed on the comment page of the content details page; and / or, if the comment keyword of the target comment content is the set keyword corresponding to the target image, a search entry is displayed on the comment page of the content details page; and / or, if the emotion type of the target comment content is positive, a search entry is displayed on the comment page of the content details page.

[0191] As an example, Figure 5b As shown, when the user is viewing the content details page of the target published content, he / she posts the target comment content "Requesting for the link to the second picture" in the comment area. At this time, the intention of the target comment content can be identified, and the intention identification result of "Requesting for the link to the second picture" is requesting for recommendation, indicating the user's search intention. At this time, the search entry "Try to search for the same item" can be displayed below the target comment content.

[0192] Step 504: In response to the triggering operation of the search entry in the comment page, a search page for the target image is displayed, and the search page includes at least one image recognition result, a summary content display area and a search result display area. The image recognition result is obtained by performing image recognition on the target image.

[0193] Using the above example, Figure 5b As shown, the user triggers the search entry "Try to search for the same item". At this time, the target image is Figure 2 in the content details page of the target published content. The search is performed based on Figure 2, and the corresponding search page is displayed. The search page can display the image recognition results "Restaurant A", "Restaurant B", and "Restaurant C" in Figure 2, as well as summary information of each image recognition result, such as "The author mentioned 3 restaurants located in XX, namely Restaurant A, Restaurant B, and Restaurant C. The average cost per person in the three restaurants is around 200, which is suitable for group gatherings to try more dishes. The atmosphere is suitable for dates", as well as the summary information block of "Restaurant A", the summary information block of "Restaurant B", and the summary information block of "Restaurant C". Triggering the summary information block can enter the corresponding related content list page, transaction page, detail page, etc.

[0194] Step 506: In response to a selection operation on a target image recognition result among the at least one image recognition result, display summary information of the target image recognition result in the summary content display area, and display search results of the target image recognition result in the search result display area.

[0195] Continuing with the above example, suppose the user selects "Restaurant A" and Restaurant A is the target image recognition result. At this time, the summary information of "Restaurant A" can be displayed in the summary content display area: "Restaurant A's special dishes include dishes A and B, the environment is XXXX, it is located at XXX location, with convenient transportation, suitable for family gatherings, and the restaurant sells surrounding tableware~" and notes related to "Restaurant A" can be displayed in the search result display area.

[0196] It should be noted that when a user views the content details page of a target published content on a content application platform, if the user is interested in a target image in the content details page of the target published content and wants to obtain relevant information about the target image, the user can leave a message in the comment area of ​​the published content to ask the publisher and then wait for the publisher to reply. This results in low efficiency in obtaining information, which greatly affects the user experience.

[0197] Therefore, one embodiment of the present specification provides a content search method. In a comment scenario, if it is monitored that a user posts target comment content on the content details page of the target published content, the target comment content can be identified and analyzed to determine whether it includes the target image, whether the intention recognition result matches the target image, whether the included comment keywords are the set keywords corresponding to the target image, and whether the emotional type is positive, thereby identifying whether the actual intention of the user to post the target comment content is to search information, and thus providing the user with a search entry on the comment page of the content details page in a timely manner. When the user triggers the search entry, a one-click search for the target image can be realized, and a search page for the target image can be displayed. At least one image recognition result, summary information, search results, etc. of the target image are displayed on the search page, and the user can be supported to flexibly select the target image recognition result of interest from the at least one identified image recognition result, and display the summary information and search results of the target image recognition result. By identifying whether the target comment content posted by the user includes the target image, whether the intent recognition result matches the target image, whether the included comment keywords are the set keywords corresponding to the target image, and whether the emotion type is positive, a search entry can be provided on the comment page when the user has a search intention, realizing one-click search based on the target comment content without waiting for the publisher's reply, thereby improving the efficiency of search information acquisition and ensuring user experience.

[0198] The above is a schematic solution of a content search method in the comment scene of this embodiment. It should be noted that the technical solution of the content search method in the comment scene is the same as the above Figure 1-Figure 3The technical solutions of the content search method shown in the figure belong to the same concept. For details not described in detail in the technical solution of the content search method in the comment scenario, please refer to the above Figure 1-Figure 3 Description of the technical solution of the content search method shown.

[0199] Figure 6 A flow chart of a content search method provided by an embodiment of this specification is shown. Figure 6 As shown, the client is an operating terminal held by a user, used to access the content application platform; the server is the server corresponding to the content application platform.

[0200] The client sends a request to the server to obtain the target published content.

[0201] The server returns the content details page of the target published content to the client.

[0202] The client displays the content details page of the target published content; in response to an interactive event on the content details page of the target published content, a search portal is displayed on the content details page. The interactive event on the content details page includes at least one of a viewing event, a comment event, and a screenshot event. The display position of the search portal corresponding to different interactive events is different.

[0203] The client, in response to the triggering operation of the search portal, sends a search event for the target image in the content details page to the server.

[0204] The server, in response to the search event, performs image recognition on the target image to obtain at least one image recognition result, generates overall summary information and at least one summary information block based on the at least one image recognition result; and generates a search page for the target image based on the at least one image recognition result, the overall summary information, and the at least one summary information block, the search page including the at least one image recognition result, a summary content display area, and a search result display area.

[0205] The server returns the search page to the client.

[0206] The client displays a search page for a target image, displays at least one image recognition result on the search page, displays overall summary information of the at least one image recognition result and at least one corresponding summary information block in a summary content display area, and displays overall search results of the at least one image recognition result in a search result display area.

[0207] The client, in response to a triggering operation of a target summary information block in at least one summary information block, sends a request for obtaining a related page corresponding to the target summary information block to the server.

[0208] The server, in response to the request to obtain the associated page, returns the associated page corresponding to the target summary information block.

[0209] The client displays the associated page corresponding to the target summary information block.

[0210] The client, in response to a selection operation on a target image recognition result in at least one image recognition result, sends the selection operation to the server.

[0211] The server generates summary information of the target image recognition result based on the target image recognition result, and searches based on the target image recognition result to obtain corresponding search results; and returns the summary information of the target image recognition result and the search results of the target image recognition result to the client.

[0212] The client displays summary information of the target image recognition result in the summary content display area and displays search results of the target image recognition result in the search result display area, wherein the summary information of the target image recognition result includes at least one related search content.

[0213] The client, in response to a triggering operation of a first related search content, displays a related content details page corresponding to the first related search content, wherein the related content details page includes at least one of a product details page, a merchant details page, a topic details page, and an activity details page; in response to a triggering operation of a second related search content, displays an account homepage corresponding to the second related search content; in response to a triggering operation of a third related search content, displays a search content list corresponding to the third related search content; in response to a triggering operation of a fourth related search content, displays a product transaction interface corresponding to the fourth related search content; and in response to a triggering operation of a fifth related search content, displays a search page for the fifth related search content.

[0214] The client, in response to the first trigger operation of the target-related search question in the search page, displays the search page of the target-related search question; in response to the second trigger operation of the target-related search question, displays the question-and-answer interaction page, and displays the search results of the target-related search question on the question-and-answer interaction page.

[0215] The client, in response to a triggering operation of a dialogue interaction entrance in a search page, displays a question-and-answer interaction page and displays at least one candidate question on the question-and-answer interaction page.

[0216] One embodiment of this specification provides a content search method that displays at least one image recognition result on a search page for selection. Upon selecting a target image recognition result, summary information and search results for the target image recognition result are displayed. The search page displays rich content, and the summary information allows users to quickly access relevant information obtained from searching for the target image, improving information acquisition efficiency and ensuring a pleasant user experience. Furthermore, the method provides a variety of interactive links that better meet users' actual search needs, allowing users to quickly obtain the search information they need, while also facilitating platform content promotion.

[0217] Corresponding to the above method embodiment, this specification also provides a content search device embodiment, Figure 7 A schematic diagram of the structure of a content search device provided by an embodiment of this specification is shown, which is applied to a client, such as Figure 7 As shown, the device includes:

[0218] The first display module 702 is configured to display a search page for a target image, the search page including at least one image recognition result, a summary content display area, and a search result display area; the image recognition result is obtained by performing image recognition on the target image;

[0219] The second display module 704 is configured to display summary information of the target image recognition result in the summary content display area and display search results of the target image recognition result in the search result display area in response to a selection operation on the target image recognition result in the at least one image recognition result.

[0220] Optionally, the first display module 702 is further configured to:

[0221] Displaying a search page for a target image, the search page including at least one image recognition result, a summary content display area, and a search result display, wherein the summary content display area displays overall summary information of the at least one image recognition result and at least one corresponding summary information block;

[0222] The device further includes a third display module configured to:

[0223] In response to a triggering operation of a target summary information block in at least one summary information block, a related page corresponding to the target summary information block is displayed.

[0224] Optionally, the search page further includes a target image; and the device further includes a fourth display module configured to:

[0225] In response to a frame selection operation on a target image, newly added image recognition results are displayed on the search page, wherein the newly added image recognition results are obtained based on recognition of the framed content corresponding to the frame selection operation.

[0226] Optionally, the summary information of the target image recognition result includes at least one associated search content; and the device further includes a fifth display module configured to do at least one of the following:

[0227] In response to a triggering operation of the first related search content, displaying a related content detail page corresponding to the first related search content, wherein the related content detail page includes at least one of a product detail page, a merchant detail page, a topic detail page, and an activity detail page;

[0228] In response to a triggering operation of the second associated search content, displaying an account homepage corresponding to the second associated search content;

[0229] In response to a triggering operation of the third associated search content, displaying a search content list corresponding to the third associated search content;

[0230] In response to a triggering operation of the fourth related search content, displaying a commodity transaction interface corresponding to the fourth related search content;

[0231] In response to the triggering operation of the fifth associated search content, a search page of the fifth associated search content is displayed.

[0232] Optionally, the device further includes a sixth display module configured to:

[0233] In response to an interactive event on a content details page of a target published content, a search entry is displayed on the content details page; the content details page includes a target image;

[0234] The first display module 702 is further configured to:

[0235] In response to a triggering operation of the search portal, a search page for the target image is displayed.

[0236] Optionally, the interactive event of the content details page includes at least one of a viewing event, a comment event, and a screenshot event; and the sixth display module is further configured to:

[0237] In response to the viewing event, displaying a search entry at the image and / or text of the content details page;

[0238] In response to a comment event, a search entry is displayed on the comment page of the content details page;

[0239] In response to the screenshot event, a search entry is displayed on the screenshot interface of the content details page.

[0240] Optionally, the sixth display module is further configured as any one of the following:

[0241] In response to publishing a target review content including a target image, displaying a search portal on a review page of the content details page;

[0242] In response to the target review content being published, performing intent recognition on the target review content, and when the result of the intent recognition matches the target image, displaying a search entry on the review page of the content details page;

[0243] In response to the target review content being published, extracting the review keywords of the target review content, and when the review keywords are set keywords corresponding to the target image, displaying a search entry on the review page of the content details page;

[0244] In response to the target review content being published, the sentiment type of the target review content is identified, and when the sentiment type is a positive type, a search entry is displayed on the review page of the content details page.

[0245] Optionally, the search page further includes at least one associated search question, which is generated based on the target image recognition result, summary information of the target image recognition result, and / or search results of the target image recognition result; the device further includes a seventh display module, which is configured to:

[0246] In response to a first triggering operation of a target-related search question, displaying a search page of the target-related search question, wherein the target-related search question is a triggered question among the at least one related search question;

[0247] In response to the second triggering operation of the target-related search question, a question-and-answer interaction page is displayed, and the search results of the target-related search question are displayed on the question-and-answer interaction page.

[0248] Optionally, the search page further includes a dialogue interaction entrance, and the device further includes an eighth display module configured to:

[0249] In response to the triggering operation of the dialogue interaction entrance, a question-and-answer interaction page is displayed, and at least one candidate question is displayed on the question-and-answer interaction page, wherein the at least one candidate question is generated based on at least one of the target user portrait, similar user portraits, information on at least one dimension of the target published content, interactive events, target image recognition results, summary information of the target image recognition results, and search results of the target image recognition results.

[0250] Optionally, the device further includes a determining module configured to:

[0251] Generate target prompt information based on the target image recognition result and the preset prompt word template;

[0252] The target prompt information is input into the target generation model to obtain the summary information of the target image recognition result.

[0253] An embodiment of the present specification provides a content search device, which can display at least one image recognition result obtained by identification on a search page for selection. When a target image recognition result is selected, summary information of the target image recognition result and search results of the target image recognition result can be displayed. The content displayed on the search page is rich, and the relevant information obtained by searching for the target image can be quickly obtained through the summary information, thereby improving information acquisition efficiency and ensuring user experience.

[0254] Corresponding to the above method embodiment, this specification also provides a content search device embodiment, Figure 8 A schematic diagram of the structure of another content search device provided by an embodiment of this specification is shown, which is applied to a server, such as Figure 8 As shown, the device includes:

[0255] A first returning module 802 is configured to, in response to a search event for a target image, perform image recognition on the target image to obtain at least one image recognition result, generate a search page for the target image, and return the search page to the client, the search page including at least one image recognition result, a summary content display area, and a search result display area;

[0256] A determination module 804 is configured to determine summary information corresponding to the target image recognition result and search results corresponding to the target image recognition result;

[0257] The second return module 806 is configured to return the summary information of the target image recognition result and the search results of the target image recognition result to the client, so that the client displays the summary information of the target image recognition result in the summary content display area and displays the search results of the target image recognition result in the search result display area.

[0258] An embodiment of the present specification provides a content search device, in which a server can perform image recognition on a target image to obtain at least one image recognition result, generate a search page for the target image, and provide at least one image recognition result for selection on the search page. When the target image recognition result is selected, the server can also generate summary information of the target image recognition result based on the target image recognition result, and perform a search based on the target image recognition result to obtain corresponding search results, and feed back rich content to the client, so that the user can quickly obtain relevant information obtained by searching the target image through the summary information, thereby improving information acquisition efficiency and ensuring user experience.

[0259] The above is a schematic solution of a content search device of this embodiment. It should be noted that the technical solution of the content search device and the technical solution of the above-mentioned content search method are of the same concept. For details not described in detail in the technical solution of the content search device, please refer to the description of the technical solution of the above-mentioned content search method.

[0260] Figure 9 1 shows a block diagram of a computing device according to one embodiment of the present disclosure. Components of the computing device 900 include, but are not limited to, a memory 910 and a processor 920. The processor 920 is connected to the memory 910 via a bus 930, and a database 950 is used to store data.

[0261] The computing device 900 also includes an access device 940 that enables the computing device 900 to communicate via one or more networks 960. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 940 may include one or more of any type of network interface (e.g., a network interface card (NIC)) whether wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, or a near field communication (NFC) interface.

[0262] In one embodiment of the present specification, the above components of the computing device 900 and Figure 9 Other components not shown in the figure may also be connected to each other, for example, via a bus. Figure 9 The computing device structure block diagram shown is for illustrative purposes only and is not intended to limit the scope of this specification. Those skilled in the art may add or replace other components as needed.

[0263] The computing device 900 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook computer, etc.), a mobile phone (e.g., a smartphone), a wearable computing device (e.g., a smartwatch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or a personal computer (PC). The computing device 900 may also be a mobile or stationary server.

[0264] The processor 920 is configured to execute the following computer-executable instructions, which implement the steps of the above-mentioned content search method when executed by the processor.

[0265] The above is a schematic solution of a computing device of this embodiment. It should be noted that the technical solution of the computing device and the technical solution of the above-mentioned content search method are based on the same concept. For details not described in detail in the technical solution of the computing device, please refer to the description of the technical solution of the above-mentioned content search method.

[0266] An embodiment of the present specification further provides a computer-readable storage medium storing computer-executable instructions, which implement the steps of the above-mentioned content search method when executed by a processor.

[0267] The above is a schematic diagram of a computer-readable storage medium according to this embodiment. It should be noted that the technical solution of this storage medium and the technical solution of the content search method described above are based on the same concept. For details not described in detail in the technical solution of the storage medium, please refer to the description of the technical solution of the content search method described above.

[0268] An embodiment of the present specification further provides a computer program, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the above-mentioned content search method.

[0269] The above is an illustrative solution of a computer program of this embodiment. It should be noted that the technical solution of the computer program and the technical solution of the above-mentioned content search method are of the same concept. For details not described in detail in the technical solution of the computer program, please refer to the description of the technical solution of the above-mentioned content search method.

[0270] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0271] Computer instructions include computer program code, which may be in source code form, object code form, executable files, or some intermediate form. Computer-readable media may include any entity or device capable of carrying computer program code, recording media, USB flash drives, mobile hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signals, telecommunications signals, and software distribution media. It should be noted that the content of computer-readable media may be appropriately increased or decreased based on the requirements of patent practice. For example, in some regions, according to patent practice, computer-readable media does not include electric carrier signals and telecommunications signals.

[0272] It should be noted that for the aforementioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of this specification are not limited by the order of the actions described, because according to the embodiments of this specification, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the embodiments of this specification.

[0273] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0274] The preferred embodiments disclosed above are intended only to help illustrate this specification. The optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations can be made based on the content of the embodiments of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the embodiments of this specification, so that those skilled in the art can better understand and utilize this specification. This specification is limited only by the claims and their full scope and equivalents.

Claims

1. A content search method, characterized in that: Applied to the client, including: Displaying a search page for a target image, the search page including at least one image recognition result, a summary content display area, and a search result display area; the image recognition result is obtained by performing image recognition on the target image; In response to a selection operation of a target image recognition result among the at least one image recognition result, summary information of the target image recognition result is displayed in the summary content display area, and search results of the target image recognition result are displayed in the search result display area.

2. The content search method according to claim 1, characterized in that: The display of a search page for a target image, the search page including at least one image recognition result, a summary content display area, and a search result display area, includes: Displaying a search page for a target image, the search page including at least one image recognition result, a summary content display area, and a search result display, wherein the summary content display area displays overall summary information of the at least one image recognition result and at least one corresponding summary information block; The method further comprises: In response to a triggering operation of a target summary information block in the at least one summary information block, a related page corresponding to the target summary information block is displayed.

3. The content search method according to claim 1, wherein: The search page also includes the target image; and the method further includes: In response to the frame selection operation of the target image, a new image recognition result is displayed on the search page, wherein the new image recognition result is obtained based on the recognition of the framed content corresponding to the frame selection operation.

4. The content search method according to claim 1, characterized in that: The summary information of the target image recognition result includes at least one associated search content; and the method further includes at least one of the following: In response to a triggering operation of the first related search content, displaying a related content details page corresponding to the first related search content, wherein the related content details page includes at least one of a product details page, a merchant details page, a topic details page, and an activity details page; In response to a triggering operation of the second associated search content, displaying an account homepage corresponding to the second associated search content; In response to a triggering operation of the third associated search content, displaying a search content list corresponding to the third associated search content; In response to a triggering operation of the fourth related search content, displaying a commodity transaction interface corresponding to the fourth related search content; In response to the triggering operation of the fifth associated search content, a search page of the fifth associated search content is displayed.

5. The content search method according to claim 1, characterized in that: Before displaying the search page for the target image, the method further includes: In response to an interactive event on a content details page of a target published content, displaying a search entry on the content details page; the content details page includes the target image; The displaying of a search page for a target image includes: In response to the triggering operation of the search portal, the search page for the target image is displayed.

6. The content search method according to claim 5, characterized in that: The interactive event of the content details page includes at least one of a viewing event, a comment event, and a screenshot event; The step of displaying a search entry on a content details page in response to an interactive event on the content details page of the target published content includes: In response to the viewing event, displaying the search entry at the image and / or text of the content details page; In response to the comment event, displaying the search entry on the comment page of the content details page; In response to the screenshot event, the search entry is displayed on the screenshot interface of the content details page.

7. The content search method according to claim 6, characterized in that: In response to the comment event, displaying the search entry on the comment page of the content details page includes any one of the following methods: In response to publishing the target review content including the target image, displaying the search entry on the review page of the content details page; In response to publishing a target review content, performing intent recognition on the target review content, and when a result of the intent recognition matches the target image, displaying the search entry on the review page of the content details page; In response to publishing a target review content, extracting a review keyword of the target review content, and displaying the search entry on the review page of the content details page when the review keyword is a set keyword corresponding to the target image; In response to publishing target review content, the sentiment type of the target review content is identified, and when the sentiment type is a positive type, the search entry is displayed on the review page of the content details page.

8. The content search method according to claim 1, characterized in that: The search page further includes at least one associated search question, where the associated search question is generated based on the target image recognition result, summary information of the target image recognition result, and / or search results of the target image recognition result; and the method further includes at least one of the following: In response to a first triggering operation of a target-related search question, displaying a search page of the target-related search question, wherein the target-related search question is a triggered question among the at least one related search question; In response to the second triggering operation of the target-related search question, a question-and-answer interaction page is displayed, and the search results of the target-related search question are displayed on the question-and-answer interaction page.

9. The content search method according to claim 1, characterized in that: The search page also includes a dialogue interaction entrance, and the method further includes: In response to the triggering operation of the dialogue interaction entrance, a question-and-answer interaction page is displayed, and at least one candidate question is displayed on the question-and-answer interaction page, wherein the at least one candidate question is generated based on at least one item of the target user portrait, similar user portraits, information of at least one dimension of the target published content, interactive events, the target image recognition result, summary information of the target image recognition result, and search results of the target image recognition result.

10. The content search method according to claim 1, characterized in that: The summary information of the target image recognition result is determined by the following method: Generate target prompt information based on the target image recognition result and the preset prompt word template; The target prompt information is input into a target generation model to obtain summary information of the target image recognition result.

11. A content search method, characterized in that: Applied to the server, including: In response to a search event for a target image, performing image recognition on the target image to obtain at least one image recognition result, generating a search page for the target image, and returning the search page to the client, the search page including the at least one image recognition result, a summary content display area, and a search result display area; Determining summary information corresponding to a target image recognition result and search results corresponding to the target image recognition result; The summary information of the target image recognition result and the search results of the target image recognition result are returned to the client, so that the client displays the summary information of the target image recognition result in the summary content display area and displays the search results of the target image recognition result in the search result display area.

12. A content search device, characterized in that: Applied to the client, including: A first display module is configured to display a search page for a target image, the search page including at least one image recognition result, a summary content display area, and a search result display area; the image recognition result is obtained by performing image recognition on the target image; The second display module is configured to display summary information of the target image recognition result in the summary content display area and display search results of the target image recognition result in the search result display area in response to a selection operation of the target image recognition result in the at least one image recognition result.

13. A content search device, characterized in that: Applied to the server, including: a first returning module configured to, in response to a search event for a target image, perform image recognition on the target image to obtain at least one image recognition result, generate a search page for the target image, and return the search page to the client, the search page including the at least one image recognition result, a summary content display area, and a search result display area; a determination module configured to determine summary information corresponding to a target image recognition result and a search result corresponding to the target image recognition result; The second returning module is configured to return the summary information of the target image recognition result and the search results of the target image recognition result to the client, so that the client displays the summary information of the target image recognition result in the summary content display area and displays the search results of the target image recognition result in the search result display area.

14. A computing device, characterized in that include: memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the content search method according to any one of claims 1 to 11 are implemented.

15. A computer-readable storage medium, characterized in that It stores computer-executable instructions, which, when executed by a processor, implement the steps of the content search method according to any one of claims 1 to 11.

16. A computer program product, characterized in that The method comprises a computer program / instruction, which implements the steps of the content search method according to any one of claims 1 to 11 when executed by a processor.

Images