Item search method, device, server, terminal and storage medium

Abstract

The present application provides an item search method, device, server, terminal and storage medium, which belongs to the field of Internet technology. The method includes: receiving an item search request sent by a terminal, the item search request carries target object features, and the target object features are features of the target object of the lost item; responding to the item search request, based on the identification information of the first video clip captured by at least two cameras in different positions, determining the second video clip where the item is lost, the first video clip contains the target object, the identification information includes the target object features and the item tag associated with the target object features, and the item tag represents the item carried by the target object in the first video clip; sending the item search result to the terminal, the item search result includes the location information of the camera that captured the second video clip. The above scheme realizes the rapid determination of the location of the lost item and improves the convenience of item search.

Description

Technical Field

[0001] The present application relates to the field of Internet technology, and in particular to an item search method, device, server, terminal and storage medium. Background Art

[0002] At present, with the improvement of living standards, people can go to shopping malls, parks and other leisure and entertainment places to play in their spare time. During the play, people usually carry some items, such as mobile phones, hats or backpacks, etc. However, items may be lost during the play, so how to find the lost items has become an urgent problem to be solved. Summary of the invention

[0003] The embodiments of the present application provide an item search method, device, server, terminal and storage medium, which can improve the convenience of item search. The technical solution is as follows:

[0004] According to a first aspect of an embodiment of the present application, a method for searching for an item is provided, the method comprising:

[0005] An item search request sent by a receiving terminal is carried by the item search request, and the target object feature is a feature of a target object of the lost item;

[0006] In response to the item search request, determining a second video segment in which the item is lost based on identification information of a first video segment captured by at least two cameras at different locations, wherein the first video segment includes the target object, the identification information includes a feature of the target object and an item tag associated with the feature of the target object, and the item tag indicates an item carried by the target object in the first video segment;

[0007] Sending an object search result to the terminal, the object search result including location information corresponding to the camera that captures the second video clip.

[0008] In a possible implementation, determining the second video segment of the lost item based on identification information of the first video segment captured by at least two cameras at different locations includes:

[0009] For every two first video segments that are acquired at adjacent times and by cameras at different positions, if there is an object tag whose identification information of the first target video segment is included in the two first video segments and whose identification information of the second target video segment is not included in the two first video segments, at least one of the first target video segment and the second target video segment is determined as the second video segment;

[0010] The acquisition time of the first target video segment is earlier than the acquisition time of the second target video segment.

[0011] In a possible implementation, the object search request further carries a target object tag, where the target object tag indicates the lost object; if there is an object tag that includes identification information of a first target video segment in the two first video segments and does not include identification information of a second target video segment in the two first video segments, determining at least one of the first target video segment and the second target video segment as the second video segment, includes:

[0012] If the target object tag is included in the identification information of the first target video segment and is not included in the identification information of the second target video segment, at least one of the first target video segment and the second target video segment is determined as the second video segment.

[0013] In a possible implementation manner, after determining at least one of the first target video segment and the second target video segment as the second video segment and before sending the item search result to the terminal, the method further includes:

[0014] When a plurality of second video segments are determined based on a plurality of first video segments, the second video segments other than the second video segment with the latest acquisition time are filtered.

[0015] In a possible implementation manner, determining at least one of the first target video segment and the second target video segment as the second video segment includes:

[0016] For every two adjacent video frames in the first target video segment, if the identification information of the first video frame of the two video frames includes the item tag, and the identification information of the second video frame of the two video frames does not include the item tag, then determining the second video frame as the target video frame where the item is lost, determining the first target video segment as the second video segment, and the acquisition time of the first video frame is earlier than the acquisition time of the second video frame;

[0017] If the identification information of the last video frame in the first target video segment includes the object tag, the second target video segment is determined as the second video segment.

[0018] In a possible implementation, the item search request further carries a time period; and the determining, based on identification information of the first video segments captured by at least two cameras at different locations, the second video segment in which the item is lost includes:

[0019] Based on the identification information of the first video segment captured by at least two cameras in different positions, a second video segment in which the item is lost and the capture time is within the time period is determined.

[0020] In a possible implementation, before determining the second video segment of the lost item based on the identification information of the first video segment captured by at least two cameras at different positions, the method further includes:

[0021] Extracting a first video segment containing the target object from videos captured by at least two cameras at different locations;

[0022] Determine identification information of the first video segment.

[0023] In a possible implementation manner, determining the identification information of the first video segment includes:

[0024] For each video frame in the first video clip, performing object recognition on the video frame to obtain an object region and object features contained in the object region;

[0025] Performing object recognition on the video frame to obtain an object region and an object label corresponding to the object contained in the object region;

[0026] When the item area is within a preset range of the object area, an association relationship is established between the object feature and the item tag.

[0027] In a possible implementation, the performing object recognition on the video frame to obtain an object region and an object label corresponding to the object contained in the object region includes:

[0028] An object recognition model is called to perform object recognition on the video frame to obtain the object region and object labels corresponding to the objects contained in the object region.

[0029] In a possible implementation manner, determining the identification information of the first video segment includes:

[0030] Extracting a video frame from the first video segment every preset number of frames;

[0031] The extracted video frame is identified to obtain the identification information.

[0032] In a possible implementation, the method further includes:

[0033] Extracting video segments containing features of the same object from videos captured by at least two cameras at different locations;

[0034] Determining identification information of the video clip;

[0035] For every two video clips that are collected at adjacent times and by cameras at different positions, if there is identification information of a third video clip whose item tag is included in the two video clips, and identification information of a fourth video clip that is not included in the two video clips, then in the identification information of at least one of the third video clip and the fourth video clip, an item disappearance tag corresponding to the item tag is added, and the collection time of the third video clip is earlier than the collection time of the fourth video clip;

[0036] The method of determining the second video segment of the lost item based on identification information of the first video segment collected by at least two cameras at different positions includes:

[0037] From the first video segments captured by at least two cameras at different positions, a second target video segment whose identification information includes the missing tag of the object is searched, and the second target video segment is determined to be the second video segment.

[0038] According to a second aspect of an embodiment of the present application, a method for searching for an item is provided, the method comprising:

[0039] Display the item search interface;

[0040] Based on the item search interface, collecting target object features corresponding to the target object;

[0041] Sending an item search request to a server, the item search request carrying the target object feature, the server being configured to respond to the item search request, determine a second video segment where the item is lost based on identification information of a first video segment captured by at least two cameras at different locations, and return an item search result, the item search result including location information corresponding to the camera that captured the second video segment;

[0042] receiving the item search result and displaying the item search result;

[0043] The first video clip contains the target object, the identification information includes the target object features and an item tag associated with the target object features, and the item tag indicates an item carried by the target object in the first video clip.

[0044] In a possible implementation, the item search request further carries a target item tag, where the target item tag indicates the lost item, and the method further includes:

[0045] The target item label input in the item search interface is obtained.

[0046] In a possible implementation, the item search request further carries a time period, and the method further includes:

[0047] The time period entered in the item search interface is obtained.

[0048] According to a third aspect of an embodiment of the present application, there is provided an object search device, the device comprising:

[0049] A request receiving module, configured to receive an item search request sent by a terminal, wherein the item search request carries a target object feature, and the target object feature is a feature of a target object of a lost item;

[0050] a video segment determination module, configured to determine, in response to the item search request, a second video segment in which the item is lost based on identification information of a first video segment captured by at least two cameras at different locations, wherein the first video segment includes the target object, the identification information includes a feature of the target object and an item tag associated with the feature of the target object, and the item tag indicates an item carried by the target object in the first video segment;

[0051] The result sending module is used to send the object search result to the terminal, and the object search result includes the position information corresponding to the camera that captures the second video clip.

[0052] In a possible implementation, the video segment determination module is configured to:

[0053] For every two first video segments that are acquired at adjacent times and by cameras at different positions, if there is an object tag whose identification information of the first target video segment is included in the two first video segments and whose identification information of the second target video segment is not included in the two first video segments, at least one of the first target video segment and the second target video segment is determined as the second video segment;

[0054] The acquisition time of the first target video segment is earlier than the acquisition time of the second target video segment.

[0055] In a possible implementation, the object search request further carries a target object tag, where the target object tag indicates the lost object; and the video segment determination module is configured to:

[0056] If the identification information of the first target video segment includes the target object tag, and the identification information of the second target video segment does not include the target object tag, at least one of the first target video segment and the second target video segment is determined as the second video segment.

[0057] In a possible implementation manner, the device further includes:

[0058] The video segment filtering module is used to filter the second video segments except the second video segment with the latest acquisition time when multiple second video segments are determined based on multiple first video segments.

[0059] In a possible implementation, the video segment determination module is configured to:

[0060] For every two adjacent video frames in the first target video segment, if the identification information of the first video frame of the two video frames includes the item tag, and the identification information of the second video frame of the two video frames does not include the item tag, then determining the second video frame as the target video frame where the item is lost, determining the first target video segment as the second video segment, and the acquisition time of the first video frame is earlier than the acquisition time of the second video frame;

[0061] If the identification information of the last video frame in the first target video segment includes the object tag, the second target video segment is determined as the second video segment.

[0062] In a possible implementation, the item search request further carries a time period; and the video segment determination module is configured to:

[0063] Based on the identification information of the first video segment captured by at least two cameras in different positions, a second video segment in which the item is lost and the capture time is within the time period is determined.

[0064] In a possible implementation manner, the device further includes:

[0065] A video segment extraction module, used to extract a first video segment containing the target object from videos captured by at least two cameras at different positions;

[0066] The identification information determination module is used to determine the identification information of the first video clip.

[0067] In a possible implementation, the identification information determination module is used to:

[0068] For each video frame in the first video clip, performing object recognition on the video frame to obtain an object region and object features contained in the object region;

[0069] Performing object recognition on the video frame to obtain an object region and an object label corresponding to the object contained in the object region;

[0070] When the item area is within a preset range of the object area, an association relationship is established between the object feature and the item tag.

[0071] In a possible implementation, the identification information determination module is used to:

[0072] An object recognition model is called to perform object recognition on the video frame to obtain the object region and object labels corresponding to the objects contained in the object region.

[0073] In a possible implementation, the identification information determination module is used to:

[0074] Extracting a video frame from the first video segment every preset number of frames;

[0075] The extracted video frame is identified to obtain the identification information.

[0076] In a possible implementation manner, the device further includes:

[0077] A video segment extraction module, used to extract video segments containing features of the same object from videos captured by at least two cameras at different locations;

[0078] An identification information determination module, used to determine the identification information of the video clip;

[0079] a label adding module, configured to, for each two video segments that are acquired at adjacent times and acquired by cameras at different positions, add an item disappearance label corresponding to the item label in the identification information of at least one of the third video segment and the fourth video segment if there is an item label included in the identification information of a third video segment in the two video segments and not included in the identification information of a fourth video segment in the two video segments, and the acquisition time of the third video segment is earlier than the acquisition time of the fourth video segment;

[0080] The video segment determination module is used to search for a second target video segment whose identification information includes the missing object tag from first video segments captured by at least two cameras at different positions, and determine that the second target video segment is the second video segment.

[0081] According to a fourth aspect of an embodiment of the present application, there is provided an object search device, the device comprising:

[0082] An interface display module, used to display an item search interface;

[0083] A feature collection module, used for collecting target object features corresponding to the target object based on the item search interface;

[0084] a request sending module, configured to send an item search request to a server, wherein the item search request carries the target object feature, and wherein the server is configured to respond to the item search request, determine a second video segment where the item is lost based on identification information of a first video segment captured by at least two cameras at different locations, and return an item search result, wherein the item search result includes location information of a camera that captured the second video segment;

[0085] A result receiving module, used for receiving the item search result and displaying the item search result;

[0086] The first video clip contains the target object, the identification information includes the target object features and an item tag associated with the target object features, and the item tag indicates an item carried by the target object in the first video clip.

[0087] In a possible implementation, the item search request further carries a target item tag, where the target item tag indicates the lost item, and the device further includes:

[0088] The label acquisition module is used to acquire the target item label input in the item search interface.

[0089] In a possible implementation, the item search request also carries a time period, and the device further includes:

[0090] The time period acquisition module is used to acquire the time period input in the item search interface.

[0091] According to a fifth aspect of an embodiment of the present application, a server is provided, comprising a processor and a memory, wherein the memory stores at least one program code, and the at least one program code is loaded and executed by the processor to implement an item search method as described in any possible implementation method provided in the first aspect above.

[0092] According to the sixth aspect of an embodiment of the present application, a terminal is provided, comprising a processor and a memory, wherein the memory stores at least one program code, and the at least one program code is loaded and executed by the processor to implement an object search method as described in any possible implementation method provided in the second aspect above.

[0093] According to the seventh aspect of an embodiment of the present application, a computer-readable storage medium is provided, in which at least one program code is stored, and the at least one program code is loaded and executed by a processor to implement the object search method described in any possible implementation manner provided in the first aspect above, or to implement the object search method described in any possible implementation manner provided in the second aspect above.

[0094] According to an eighth aspect of an embodiment of the present application, a computer program product is provided, the computer program product comprising a computer program code, the computer program code being stored in a computer-readable storage medium, a processor reading the computer program code from the computer-readable storage medium, the processor executing the computer program code to implement the object search method described in any possible implementation manner provided in the first aspect above, or to implement the object search method described in any possible implementation manner provided in the second aspect above.

[0095] An embodiment of the present application provides a solution for locating the location of a lost item after it is lost. When searching for the lost item of a target object, for a first video clip captured by at least two cameras in different positions and containing the target object, since the identification information of the first video clip can indicate which items the target object carried in the first video clip, based on the identification information of the first video clip, the second video clip where the item is lost can be further located, thereby obtaining an item search result. The location indicated by the location information included in the item search result is likely to be the location where the item is lost, so that the target object who lost the item can find the item based on the location information, thereby achieving rapid determination of the location where the item is lost and improving the convenience of item search. BRIEF DESCRIPTION OF THE DRAWINGS

[0096] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0097] Figure 1 It is a structural diagram of an implementation environment provided by an embodiment of the present application;

[0098] Figure 2 is a flow chart of an item search method provided by an embodiment of the present application;

[0099] Figure 3 is a flow chart of another method for searching for items provided in an embodiment of the present application;

[0100] Figure 4 is a flow chart of another method for searching for items provided in an embodiment of the present application;

[0101] Figure 5 is a flow chart of another method for searching for items provided in an embodiment of the present application;

[0102] Figure 6 is a flow chart of another method for searching for items provided in an embodiment of the present application;

[0103] Figure 7 is a structural schematic diagram of an object search device provided in an embodiment of the present application;

[0104] Figure 8 is a schematic diagram of the structure of another object search device provided in an embodiment of the present application;

[0105] Fig. 9 It is a structural diagram of a server provided in an embodiment of the present application;

[0106] Fig.10 It is a structural diagram of a terminal provided in an embodiment of the present application. DETAILED DESCRIPTION

[0107] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0108] It can be understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept.

[0109] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0110] It should be noted that the information (including but not limited to identification information, location information, etc.) and data (including but not limited to data for processing, data for storage, data for display, etc.) involved in this application are all 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 the relevant countries and regions. For example, the videos involved in this application were obtained with full authorization.

[0111] Figure 1 This is a schematic diagram of the structure of an implementation environment provided by the embodiment of the present application, see Figure 1 The implementation environment includes a terminal 101, a server 102 and at least one camera 103. The terminal 101 and each camera 103 are respectively connected to the server 102 via a wireless or wired network.

[0112] Optionally, the terminal 101 includes but is not limited to a smart phone, a tablet computer, a laptop computer or a desktop computer. Optionally, the server 102 is at least one of a server, a server cluster consisting of multiple servers, a cloud server, a cloud computing platform and a virtualization center, which is not limited in the embodiments of the present application. The camera 103 is used to collect video. The camera 103 is set in a place where items may be lost, such as a shopping mall, a park or a station.

[0113] In the embodiment of the present application, each camera 103 is set at a different position, and the camera 103 is used to collect videos at the corresponding position and send the videos to the server 102. The server 102 is used to manage the camera 103 and the videos collected by the camera 103. Accordingly, the server 102 is used to receive the videos sent by the camera 103 and store the videos. Optionally, the server 102 sets a corresponding video file for the camera 103, and the video file stores the video collected by the camera 103.

[0114] In the embodiment of the present application, the server 102 provides an item search service to the outside world, and the terminal 101 searches for items by accessing the item search service provided by the server 102. Optionally, the terminal 101 is a control terminal for managing the server 102, and is used by the administrator of the server 102, so that the administrator can access the server 102 through the terminal 101. Alternatively, the terminal 101 is a terminal used by any object, and any object can access the server 102 through its own terminal 101.

[0115] Optionally, the server 102 provides a web page to the outside, and the terminal 101 implements functions such as item search, image acquisition, or information display by accessing the web page. Exemplarily, the terminal 101 can access the web page in a browser or any application that contains a webview (web view) control. Alternatively, a target application served by the server 102 is installed on the terminal 101, and the terminal 101 can implement functions such as item search, image acquisition, and information display through the target application. Optionally, the target application is a target application in the operating system of the terminal 101, or a target application provided by a third party. For example, the target application is an item search application or a shopping application.

[0116] In the embodiment of the present application, after losing an item, the subject can access the item search service provided by the server 102 through the terminal 101, and the server 102 returns the item search result, so that the terminal 101 receives the item search result and displays it for the subject who lost the item to view, so that the subject can find the lost item according to the item search result.

[0117] The item search method provided in the embodiment of the present application can be applied to the scenario of searching for items after the items are lost. The application scenario of the item search method is introduced below. For example, cameras are set up at different locations in the mall, such as the elevator door and the restaurant door on the fifth floor. The camera collects the video at the corresponding location and sends the video to the server. The server receives the video and stores it. Users carrying items are active in the mall. If the user loses an item, he can go to the management office of the mall. The terminal of the management office is connected to the camera. The terminal controls the camera to collect the user's facial features and sends an item search request carrying the facial features to the server. The server determines the location information of the lost item through the item search method provided in the embodiment of the present application, returns the location information, and the terminal displays the location information, so that the user can find the lost item at the location indicated by the location information, or recall the details when the item was lost, thereby improving the convenience of item search.

[0118] It should be noted that the above application scenarios are only exemplary and do not limit the item search scenarios. In addition to being applied to the above scenarios, the present application can also be applied to any other item search scenario.

[0119] Figure 2 This is a flowchart of an item search method provided in an embodiment of the present application. This embodiment of the present application is described by taking a server as the execution subject as an example. Figure 2 , the method comprises the following steps:

[0120] 201. The server receives an item search request sent by a terminal. The item search request carries a target object feature, which is a feature of a target object of a lost item.

[0121] The object feature corresponding to the object is used to describe the characteristics of the object, and the object feature is used to distinguish different objects. Optionally, the object includes a person, an animal or a thing, and accordingly, when the object is a person, the object feature is a face feature.

[0122] The terminal is a terminal corresponding to the target object of the lost item. In the embodiment of the present application, if an object loses an item, it can trigger the terminal to send an item search request to the server, and the server obtains the target object feature carried by the item search request. The target object feature is a feature corresponding to the target object, and the target object feature can indicate which object is the target object of the lost item.

[0123] 202. In response to the item search request, the server determines a second video segment where the item is lost based on identification information of a first video segment captured by at least two cameras at different locations, wherein the first video segment contains the target object, the identification information includes features of the target object and an item tag associated with the features of the target object, and the item tag indicates an item carried by the target object in the first video segment.

[0124] In the embodiment of the present application, the camera captures a video, sends the video to the server, and the server receives and stores the video. The camera is set at any position in the venue, and different cameras are located at different positions. For a camera at any position, if an object is within the camera's acquisition range, the video captured by the camera includes the object. If the object carries an item and the item is also within the camera's acquisition range, the video also includes the item.

[0125] It should be noted that the camera can send a video frame to the server every time it collects a video frame, so that the real-time performance of data transmission and reception is higher, or the camera can continuously collect multiple video frames and send the multiple video frames to the server at one time, thereby reducing the number of data transmission and reception times. The embodiments of the present application are not limited to this.

[0126] After receiving the object search request, the server obtains identification information of the first video clip captured by at least two cameras at different locations. The identification information of the video clip includes object features and object tags associated with the object features, and the object features included in the identification information are object features corresponding to the object included in the video clip. The object tags associated with the object features are item tags corresponding to the items carried by the object in the video clip. Optionally, the item tags include item categories, such as the item tags are mobile phones, backpacks, or cups. Optionally, the item tags also include item colors, such as black, white, or red.

[0127] Wherein, if the first video clip contains the target object, the identification information includes the target object features corresponding to the target object, and if the first video clip also contains items carried by the target object, the identification information also includes the item tag associated with the target object features, and the item tag is used to represent the items carried by the target object to which the target object features belong. The first video clip containing the target object is a video clip where the target object whose item is lost appears. The location information of the camera that captures the first video clip can indicate the locations where the target object has appeared. Moreover, since the server identifies the item tag associated with the target object features for the first video clip, the second video clip where the item is lost can be further determined based on the item tag.

[0128] 203. The server sends an object search result to the terminal, where the object search result includes location information corresponding to the camera that captures the second video clip.

[0129] The position information corresponding to the camera that captured the second video clip indicates the position of the camera. In the embodiment of the present application, the position of the camera that captured the second video clip is likely to be the position where the item is lost. By returning the position information corresponding to the camera that captured the second video clip as the item search result to the terminal, the terminal can display the item search result for the target object to view, so that the target object can know the position where the item is lost and find the item according to the position.

[0130] An embodiment of the present application provides a solution for locating the location of a lost item after it is lost. When searching for the lost item of a target object, for a first video clip captured by at least two cameras in different positions and containing the target object, since the identification information of the first video clip can indicate which items the target object carried in the first video clip, based on the identification information of the first video clip, the second video clip where the item is lost can be further located, thereby obtaining an item search result. The location indicated by the location information included in the item search result is likely to be the location where the item is lost, so that the target object who lost the item can find the item based on the location information, thereby achieving rapid determination of the location where the item is lost and improving the convenience of item search.

[0131] Figure 3 This is a flowchart of another method for searching items provided in an embodiment of the present application. This embodiment of the present application is described by taking the server as the execution subject as an example. Figure 3 , the method comprises the following steps:

[0132] 301. The server receives an item search request sent by a terminal. The item search request carries a target object feature, which is a feature of a target object of a lost item.

[0133] In an embodiment of the present application, if the target object loses an item, the terminal may be triggered to collect the target object features corresponding to the target object. Optionally, the terminal is connected to a camera. The target object is within the collection range of the camera, the terminal controls the camera to collect an image, and then performs object recognition on the image to obtain the target object features. The image is collected after authorization by the target object. After collecting the target object features, the terminal sends an item search request carrying the target object features to the server, and the server receives the item search request and obtains the target object features carried by the item search request.

[0134] It should be noted that the way in which the camera collects images is the same as the way in which the camera in the above-mentioned implementation environment collects videos, which will not be repeated here.

[0135] 302. In response to the object search request, the server extracts a first video segment containing a target object from videos captured by at least two cameras at different locations.

[0136] Different cameras are located at different positions in the venue. After the target object enters the venue, it will be captured by the camera set at its location. The video captured by the camera may contain the target object, and the first video clip containing the target object can be extracted from the captured video.

[0137] Optionally, for each video captured by each camera, object recognition is performed on each video frame in the video to obtain an object recognition result, which indicates whether the target object is recognized or not. Recognizing the target object means recognizing the target object features from the video frame. Since the item search request carries the target object features, and the target object features are the features of the target object, if the target object features are recognized from the video frame, the object recognition result is that the target object is recognized, indicating that the video frame contains the target object.

[0138] Among them, object recognition for a video frame refers to identifying the area where the object is located from the video frame, that is, the object area, and determining the image contained in the object area as the object feature, or extracting the object feature from the image contained in the object area. For example, if the object is a person, the face area is extracted from the image contained in the object area, and the image contained in the face area is determined as the object feature.

[0139] After object recognition is performed on the video captured by the camera, for each camera, the video frames with adjacent acquisition times that recognize the target object are spliced ​​to obtain the first video clip captured by the camera and containing the target object. Similarly, the first video clip corresponding to each camera is obtained. The acquisition time is the time when the camera captures the video frame. Optionally, when the camera sends the video frame to the server, it sends the acquisition time of the video frame to the server, and the server receives the acquisition time and stores the acquisition time for subsequent use.

[0140] When the target object stays or moves in a certain place, multiple cameras may capture the target object. Therefore, multiple first video clips can be extracted from the videos captured by multiple cameras. Each first video clip has a corresponding time period. The multiple first video clips are arranged in the order of the time periods. The sequence of the first video clips obtained after the arrangement can reflect the movement trajectory of the target object.

[0141] For example, when the target object is moving in a venue, it may first enter the acquisition range of camera 1 and be captured by camera 1, then enter the acquisition range of camera 2 and be captured by camera 3, and then enter the acquisition range of camera 1 again and be captured by camera 1 again. In this case, the target object does not appear continuously in the video captured by camera 1, and the video shot by camera 1 includes two first video clips, but the acquisition time of these two first video clips is not continuous, and the first video clip in the video shot by camera 2 is located between the above two first video clips.

[0142] For another example, the target object first enters the acquisition range of camera 1 and is acquired by camera 1, then the target object moves to a range that cannot be acquired by any camera, and then the target object enters the acquisition range of camera 1 again. In this case, the target object does not appear continuously in the video acquired by camera 1. The video shot by camera 1 includes two first video clips, but since no other camera acquires the target object during this period, the two first video clips can still be regarded as adjacent first video clips for subsequent processing.

[0143] It should be noted that the way the server recognizes objects in video frames is similar to the way the terminal recognizes objects in images in the above step 301. Therefore, for the same object, the target object features collected by the terminal when triggering an item search request are closer to the target object features in the identification information determined by the server, which facilitates the server to subsequently extract the first video clip containing the target object.

[0144] 303. The server determines identification information of the first video clip, where the identification information includes a target object feature and an item tag associated with the target object feature, where the item tag indicates an item carried by the target object in the first video clip.

[0145] After extracting the first video segment containing the target object, determining identification information of the first video segment. In a possible implementation, determining the identification information of the first video segment includes: for each video frame in the first video segment, performing object recognition on the video frame to obtain an object region and object features contained in the object region; performing object recognition on the video frame to obtain an object region and an object label corresponding to the object contained in the object region; and establishing an association relationship between the object feature and the object label when the object region is within a preset range of the object region.

[0146] Among them, the object area refers to the area where the object is located in the video frame. The object feature is the image contained in the object area or the feature extracted from the image. Optionally, the object area is represented by the coordinate area of ​​the object in the video frame. For example, in the case where the object is a person, the object area can be a human body area, then the object feature is a human body feature, and the object area can also be a face area, then the object feature is a face feature. Similarly, the item area refers to the area where the item is located in the video frame, optionally, the item area is represented by the coordinate area of ​​the item in the video frame. Among them, the size of the preset range can be set as needed, and the embodiment of the present application is not limited to this.

[0147] Since there may be multiple objects in a video frame, and each object may carry items, an association relationship can be established between the object features and the item tags to distinguish the items carried by different objects. Accordingly, the server can determine whether to associate the item tag with the object feature based on whether the item area is within the preset range of the object area. If the item area is within the preset range of the object area, it means that the items contained in the item area are closer to the objects contained in the object area, and the item is likely to be an item carried by the object. By establishing an association relationship between the object feature and the item tag, it is indicated that the item represented by the item tag is carried by the object to which the object feature belongs; if the item area is not within the preset range of the object area, it means that the items contained in the item area are far away from the objects contained in the object area, and the item may not be an item carried by the object, and there is no need to establish an association relationship between the object feature and the item tag. The association relationship established in this way is more accurate.

[0148] The identification information of the video frame includes the object features identified from the video frame and the item tag associated with the object features. The server combines the identification information of each video frame in the first video segment into the identification information of the first video segment.

[0149] In the embodiment of the present application, since the first video clip is an extracted video clip containing the target object feature, the identified object feature includes the target object feature, and accordingly, the identified item tag includes the item tag associated with the target object feature. The identification information determined by the above optional implementation includes not only the target object feature and the item tag associated with the target object feature, but also includes other identified object features and the item tag associated with the object feature, so that the identification information is more complete.

[0150] One point that needs to be explained is that the server may first perform object recognition on the video frames and then perform item recognition on the video frames, or may first perform item recognition on the video frames and then perform object recognition on the video frames, or may perform object recognition and item recognition on the video frames at the same time. The embodiments of the present application do not limit the order in which the server performs object recognition and item recognition on the video frames.

[0151] In another possible implementation, the purpose of the server determining the identification information of the first video clip is to provide data support for the subsequent determination of the video clip of the lost item, and the first video clip has been subjected to object identification in step 302. In the object identification process, the server may first store the identified target object region and the target object features contained in the object region, so that in step 303, the item identification may be performed only for the target object features. Accordingly, the implementation of determining the identification information of the first video clip may also include: for each video frame in the first video clip, performing item identification on the video frame to obtain the item region and the item label corresponding to the item contained in the item region; when the item region is within the preset range of the target object region, establishing an association relationship between the target object feature and the item label. This implementation does not require repeated object identification of the video frame, and the determined identification information only includes the target object feature and the item label associated with the target object feature, thereby greatly reducing the identification workload and improving the identification efficiency.

[0152] Optionally, the object label includes an object category, for example, the object label is a mobile phone, a backpack or a cup, etc. Optionally, the server performs object recognition on the video frame with the aid of an object recognition model, and accordingly, performs object recognition on the video frame to obtain the object region and the object labels corresponding to the object contained in the object region. The implementation method includes: calling the object recognition model, performing object recognition on the video frame, and obtaining the object region and the object labels corresponding to the object contained in the object region.

[0153] The object recognition model is used to recognize the input image, output the object area and the object label, and the object label includes the object category. In the embodiment of the present application, the object recognition model is obtained by training with a large number of training samples. The training samples include sample video frames and sample labels corresponding to the sample video frames, and the sample labels include sample object areas in the sample video frames and sample object labels corresponding to the objects contained in the sample object areas. Optionally, the sample labels corresponding to the sample video frames are obtained by technicians manually annotating the sample video frames.

[0154] In the embodiment of the present application, since the object recognition model is obtained through training with a large number of training samples, the recognition accuracy of the object recognition model is relatively high. By calling the object recognition model to perform object recognition on the video frame, the accuracy of the recognized object area and object label is also relatively high.

[0155] Optionally, the item tag also includes the item color, and the server performs color recognition on the identified item area to obtain the item color. Accordingly, the server combines the item color and the item category into an item tag. Since the item category and item color included in the item tag can represent the same item from different angles, the item tag can more accurately represent the item.

[0156] Another point to be explained is that the server can identify each video frame in the first video clip, or extract some video frames from the first video clip for identification, which is not limited in the embodiment of the present application. Optionally, the implementation method of determining the identification information of the first video clip includes: extracting a video frame from the first video clip every preset number of frames; identifying the extracted video frames to obtain identification information. The preset number of frames can be set as needed, for example, the preset number of frames is 5, 10 or 20, etc., which is not limited in the embodiment of the present application.

[0157] In the embodiment of the present application, since a video frame can be extracted every preset number of frames and the video frame can be identified, the server does not need to identify each video frame, thereby greatly reducing the workload of identifying video frames and improving recognition efficiency.

[0158] The embodiment of the present application takes a first video clip as an example to illustrate the process of the server determining the identification information of the first video clip. The number of extracted first video clips may be large, and the server determines the identification information of each first video clip and stores the determined identification information for subsequent item search. Optionally, the first video clips have their own clip identifiers, and the clip identifiers are used to distinguish different video clips. The server stores the clip identifiers and the identification information in correspondence. Among them, the clip identifiers of the first video clips can be set as needed, and the embodiment of the present application does not limit this.

[0159] 304. The server determines a second video segment of the lost item based on identification information of the first video segment captured by at least two cameras at different locations.

[0160] Among them, the first video clip containing the target object is a video clip in which the target object of the lost item appears. The location information of the camera that captures the first video clip can indicate the locations where the target object has appeared. Moreover, since the server recognizes the object tag associated with the target object feature in the first video clip, the second video clip in which the item is lost can be further determined based on the object tag.

[0161] Optionally, the implementation method of step 304 includes: for every two first video clips that are acquired at adjacent times and by cameras at different positions, if there is an object tag that contains identification information of the first target video clip in the two first video clips, and does not contain identification information of the second target video clip in the two first video clips, then at least one of the first target video clip and the second target video clip is determined as the second video clip, wherein the acquisition time of the first target video clip is earlier than the acquisition time of the second target video clip.

[0162] Since the first target video clip is the previous video clip of the second target video clip, the first target video clip is captured before the second target video clip. If the target object in the first target video clip carries an item, the identification information of the first target video clip includes the item tag corresponding to the item. However, if the identification information of the second target video clip does not include the item tag, indicating that the second target video clip does not include the item, the target object in the second target video clip is very likely not carrying the item, and it can be regarded as the target object has lost the item. The second target video clip and the first target video clip may both be video clips captured when the object lost the item. Therefore, at least one of the first target video clip and the second target video clip can be determined as the second video clip.

[0163] Optionally, after the target object loses an item, if he wants to find the item through the terminal, he can also input the target item tag in the terminal, and the target item tag represents the lost item, so that the item search request also carries the target item tag; accordingly, if the identification information of the first target video clip includes the target item tag, and the identification information of the second target video clip does not include the target item tag, at least one of the first target video clip and the second target video clip is determined as the second video clip.

[0164] Among them, the fact that the identification information of the second target video clip does not include the object tag can only indicate that there is a phenomenon of object loss in the second target video clip, and the object lost in the second target video clip may not be the same as the object that the target object is looking for. In the embodiment of the present application, when the object search request also carries the target object tag, since the target object tag can represent the lost object, that is, the object being searched, the second video clip determined based on the target object tag is more accurate.

[0165] In the embodiment of the present application, the number of first video segments may be large, and the number of determined second video segments may be large, so the determined second video segments may be filtered. Optionally, after determining at least one of the first target video segment and the second target video segment as the second video segment, the server, in the case of determining multiple second video segments based on multiple first video segments, filters the other second video segments except the second video segment with the latest acquisition time.

[0166] Since the target object may place the object in a place that is not easily captured by the camera when carrying the object, some of the first video clips may not include the object, and the identification information of the first video clip does not include the object tag corresponding to the object. However, in fact, the object is not lost, but is not captured, which leads to errors in the identification information of the first video clip. If the target object subsequently exposes the object within the camera's capture range, the identification information of the subsequent first video clip will again include the object tag corresponding to the object.

[0167] In an embodiment of the present application, if the second video clip with the latest acquisition time is the last video clip in which the object appears, then the second video clip is likely to be the second video clip in which the object is lost. By filtering the other second video clips except the second video clip with the latest acquisition time, the determined second video clip is more accurate.

[0168] In an embodiment of the present application, both the first target video clip and the second target video clip may be video clips where an item is lost. The first target video clip and the second target video clip may be further compared to select one video clip therefrom as the second video clip. Accordingly, the implementation method of determining at least one of the first target video clip and the second target video clip as the second video clip includes: for every two adjacent video frames in the first target video clip, if the identification information of the first video frame of the two video frames includes an item tag, and the identification information of the second video frame of the two video frames does not include an item tag, then the second video frame is determined to be the target video frame where the item is lost, and the first target video clip is determined to be the second video clip, and the acquisition time of the first video frame is earlier than the acquisition time of the second video frame; if the identification information of the last video frame in the first target video clip includes an item tag, then the second target video clip is determined to be the second video clip.

[0169] Among them, since the first video frame is acquired before the second video frame, if the target object in the first video frame carries an item, the identification information of the first video frame includes the item tag corresponding to the item, but if the identification information of the second video frame does not include the item tag corresponding to the item, it means that the second video frame does not include the item, and the target object in the second video frame is likely not to carry the item, and it can be regarded as the target object has lost the item, and the second video frame is likely to be the video frame acquired when the target object has lost the item, and the first target video segment can be regarded as the second video segment where the item is lost. However, if the identification information of the last video frame in the first target video segment includes the item tag, it means that the item appears in the last video frame, that is, the target object still carries the item, and the item is likely to be lost when the second target video segment is acquired or in the interval between the acquisition of the first target video segment and the second target video segment, and the second target video segment can be regarded as the second video segment where the item is lost.

[0170] In a possible implementation, after the target object loses the item, it can know the approximate time when the item is lost. If the item is to be searched through the terminal, the time period can also be entered in the terminal. The item search request can also carry the time period. The server determines the second video segment where the item is lost based on the identification information of the first video segment captured by at least two cameras in different positions. The implementation method also includes: determining the second video segment where the item is lost and the capture time is within the time period based on the identification information of the first video segment captured by at least two cameras in different positions.

[0171] In an embodiment of the present application, the time period carried by the item search request is the time period input by the target object, and the time period is the time period in which the target object determines that the item is lost. The server can directly determine the second video segment in which the item is lost and the acquisition time is within the time period in the first video segment, thereby narrowing the scope and making the determined second video segment more accurate.

[0172] 305. The server sends an object search result to the terminal, where the object search result includes location information corresponding to the camera that captures the second video clip.

[0173] Among them, the position information corresponding to the camera that collects the second video clip indicates the position of the camera. In the embodiment of the present application, the position of the camera that collects the second video clip is likely to be the position where the item is lost. By using the position information corresponding to the camera that collects the second video clip as the item search result and returning it to the terminal, the terminal can display the item search result for the target object to view, so that the target object can know the location where the item is lost and find the item according to the location. Optionally, after extracting each first video clip, the server stores the segment identifier of the first video clip and the position information corresponding to the camera that collects the first video clip in correspondence, so that after the second video clip is determined, the location information stored corresponding to the segment identifier is obtained according to the segment identifier of the second video clip. Optionally, the item search result also includes the acquisition time of the second video clip, so that the target object can know the approximate time when the item is lost.

[0174] An embodiment of the present application provides a solution for locating the location of a lost item after it is lost. When searching for the lost item of a target object, for a first video clip captured by at least two cameras in different positions and containing the target object, since the identification information of the first video clip can indicate which items the target object carried in the first video clip, based on the identification information of the first video clip, the second video clip where the item is lost can be further located, thereby obtaining an item search result. The location information included in the item search result is likely to be the location where the item is lost, so that the target object who lost the item can find the item based on the location information, thereby achieving rapid determination of the location where the item is lost and improving the convenience of item search.

[0175] Figure 4 This is a flowchart of another method for searching items provided in the embodiment of the present application. Figure 4 , the method comprises the following steps:

[0176] 401. The server extracts video segments containing the same object from videos captured by at least two cameras at different locations.

[0177] The implementation method of step 401 refers to the implementation method of step 302 above, which will not be repeated here.

[0178] It should be noted that the embodiment of the present application is to extract video clips containing the same object. For the first video frame in the collected video, the video frame is subjected to object recognition to obtain at least one object feature. Each object feature refers to an object included in the video frame. For the object referred to by each object feature, the server extracts the video clip containing the object from the collected video, and so on until the video clip of each object that has appeared in the video is extracted. Among them, the first video frame refers to the video frame with the earliest collection time.

[0179] 402. The server determines identification information of the video clip.

[0180] The implementation of step 402 refers to the implementation of step 303 above, which will not be described in detail here.

[0181] 403. For every two video clips that are collected at adjacent times and by cameras at different positions, if there is identification information of a third video clip whose object tag is included in the two video clips, and identification information of a fourth video clip that is not included in the two video clips, then the server adds an object disappearance label corresponding to the object label to the identification information of at least one of the third video clip and the fourth video clip, and the collection time of the third video clip is earlier than the collection time of the fourth video clip.

[0182] Optionally, the item disappearance tag includes an item tag and a disappearance keyword corresponding to the disappeared item. The representation of the item disappearance tag can be set as needed. For example, if the item tag is a backpack and the disappearance keyword is "disappeared", the item disappearance tag corresponding to the item tag is backpack disappearance, which is not limited in the present embodiment.

[0183] For example, if the identification information of the third video clip includes an item tag of a mobile phone, and the identification information of the fourth video clip includes an item tag of a mobile phone, there is no need to add an item disappeared tag; for another example, if the identification information of the third video clip includes an item tag of a mobile phone and a backpack, and the identification information of the fourth video clip includes an item tag of a mobile phone, the backpack is not included in the identification information of the fourth video clip, then an item disappeared tag corresponding to the backpack is added.

[0184] The implementation method of step 403 refers to the implementation method of step 304 above, which will not be described again.

[0185] 404. The server receives an item search request sent by the terminal. The item search request carries a target object feature, which is a feature of a target object of the lost item.

[0186] The implementation of step 404 refers to the implementation of step 301 above, which will not be described in detail here.

[0187] 405. In response to the object search request, the server searches for a second target video segment whose identification information includes a label of a missing object from first video segments captured by at least two cameras at different locations, and determines that the second target video segment is the second video segment, and the first video segment contains the target object.

[0188] Among them, if the identification information includes an object disappearance label, it means that the second target video clip has an object lost compared to the first target video clip, and the second target video clip is likely to be a video clip captured when the target object loses the object, so the video frame can be determined as the second video clip.

[0189] 406. The server sends an object search result to the terminal, where the object search result includes location information corresponding to the camera that captures the second video clip.

[0190] The implementation of step 406 refers to the implementation of step 305 above, which will not be described in detail here.

[0191] The embodiment of the present application provides a solution for locating the lost location of an item after the item is lost, and the embodiment of the present application is different from the above Figure 3 The difference between the illustrated embodiments is that, before receiving an item search request, the embodiment of the present application extracts a video clip containing the same object, and adds an item disappearance tag to the identification information of the video clip in which the item is lost, thereby realizing preprocessing of the video. Therefore, when receiving an item search request, the video clip whose identification information includes the item disappearance tag can be directly searched, and the second video clip in which the item is lost can be located more quickly, thereby obtaining an item search result. The location indicated by the location information included in the item search result is likely to be the location where the item is lost. Therefore, the target object of the lost item can find the item according to the location information, thereby realizing rapid determination of the location where the item is lost and improving the convenience of item search.

[0192] Figure 5 This is a flowchart of another method for searching items provided in the embodiment of the present application. Figure 5 , the method comprises the following steps:

[0193] 501. The terminal displays an item search interface.

[0194] Among them, when the target object wants to find the lost item, the terminal is triggered to display the item search interface. Optionally, the item search interface is an interface in the target application. The target object triggers the terminal to run the target application, thereby triggering the terminal to display the item search interface in the target application. For example, the target application is a browser, the item search interface is the front-end interactive interface of the website, and the server is the background server of the website. The target object triggers the terminal to run the browser, enters the website address in the browser, triggers the terminal to access the address, and triggers the terminal to display the item search interface.

[0195] 502. The terminal collects target object features corresponding to the target object based on the item search interface.

[0196] Optionally, a feature collection control is displayed in the item search interface, and the target object triggers the feature collection control. In response to the feature collection control being triggered, the terminal collects the target object features corresponding to the target object. Correspondingly, the terminal is connected to a camera, and the target object is within the collection range of the camera. In response to the feature collection control being triggered, the terminal controls the camera to collect images, performs object recognition on the image, and obtains the target object features. It should be noted that the image is collected after authorization by the target object.

[0197] 503. The terminal sends an item search request to the server, where the item search request carries the target object feature. The server is configured to respond to the item search request, determine the second video segment where the item is lost based on identification information of the first video segment captured by at least two cameras at different locations, and return an item search result, where the item search result includes location information corresponding to the camera that captured the second video segment.

[0198] The first video clip includes a target object, the identification information includes a target object feature and an item tag associated with the target object feature, and the item tag indicates an item carried by the target object in the first video clip. In an embodiment of the present application, the terminal sends an item search request to the server, implements the item search with the help of the server, and obtains an item search result.

[0199] 504. The terminal receives the item search result and displays the item search result.

[0200] In the embodiment of the present application, after receiving the item search result, the terminal displays the item search result. Optionally, the terminal displays the item search result in an item search interface for the object to view.

[0201] The embodiment of the present application provides a solution for locating the location of a lost item after it is lost. When the target object needs to find the item, it can trigger the terminal to send an item search request to the server, so that the lost item can be found with the help of the server. The server can locate the second video clip of the lost item based on the target object characteristics corresponding to the target object of the lost item, so as to obtain the item search result. The location indicated by the location information included in the item search result is likely to be the location where the item is lost, so that the target object of the lost item can find the item based on the location information, thereby realizing the rapid determination of the location where the item is lost and improving the convenience of item search.

[0202] Figure 6 A flowchart of another method for searching for items provided in an embodiment of the present application. Figure 6 , the method comprises the following steps:

[0203] 601. The terminal displays an item search interface.

[0204] 602. The terminal collects target object features corresponding to the target object based on the item search interface.

[0205] 603. The terminal sends an item search request to the server, where the item search request carries the target object feature.

[0206] The implementation of steps 601 to 603 refers to the implementation of steps 501 to 503, which will not be described in detail here.

[0207] Optionally, after the target object loses an item, if he wants to find the item through the terminal, he can also enter the item tag corresponding to the lost item in the item search interface, that is, the target item tag. Before step 603, the item search method provided in the embodiment of the present application also includes: the terminal obtains the target item tag entered in the item search interface; so that the item search request sent by the terminal to the server also carries the target item tag. In a possible implementation, the item search interface includes an input area, and the target object enters the target item tag corresponding to the item to be found in the input area, and the terminal obtains the target item tag entered in the input area.

[0208] In an embodiment of the present application, by obtaining the input target item tag, an item search request including the target item tag is sent to the server, so that the server can further narrow the search scope based on the target object characteristics, so that the accuracy of the determined item search result is higher.

[0209] Optionally, after the target object loses the item, it can know the approximate time when the item is lost. If the object is to be searched through the terminal, the time period can also be entered in the item search interface. Before step 603, the item search method provided in the embodiment of the present application also includes: the terminal obtains the time period entered in the item search interface; so that the item search request sent by the terminal to the server also carries the time period. In a possible implementation, the item search interface includes an input area, and the target object enters the time period in the input area, and the terminal obtains the input time period.

[0210] In an embodiment of the present application, by obtaining an input time period, an object search request also including the time period is sent to the server, so that the server can further narrow the search scope based on the characteristics of the target object and search for the second video clip whose acquisition time is within the time period, so that the accuracy of the determined object search result is higher.

[0211] 604. The server receives the item search request.

[0212] 605. The server determines a second video segment of the lost item based on identification information of the first video segment captured by at least two cameras at different locations.

[0213] The first video clip contains a target object, and the identification information includes features of the target object and an item tag associated with the features of the target object, and the item tag indicates an item carried by the target object in the first video clip.

[0214] 606. The server sends an object search result to the terminal, where the object search result includes location information corresponding to the camera that captures the second video clip.

[0215] The implementation of step 604 to step 606 refers to the implementation of step 301 to step 305, or the implementation of step 604 to step 606 refers to the implementation of step 404 to step 406, which will not be repeated here.

[0216] 607. The terminal receives the item search result and displays the item search result.

[0217] In the embodiment of the present application, after receiving the item search result, the terminal displays the item search result. Optionally, the terminal displays the item search result in an item search interface for the object to view.

[0218] The embodiment of the present application provides a solution for locating the location of a lost item after it is lost. When the target object needs to find the item, it can trigger the terminal to send an item search request to the server, so that the lost item can be found with the help of the server. The server can locate the second video clip of the lost item based on the target object characteristics corresponding to the target object of the lost item, so as to obtain the item search result. The location indicated by the location information included in the item search result is likely to be the location where the item is lost, so that the target object of the lost item can find the item based on the location information, thereby realizing the rapid determination of the location where the item is lost and improving the convenience of item search.

[0219] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present application, which will not be described one by one here.

[0220] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0221] Figure 7 is a schematic diagram of the structure of an object search device provided in an embodiment of the present application. Figure 7 , the device comprises:

[0222] The request receiving module 701 is used to receive an item search request sent by a terminal, where the item search request carries a target object feature, where the target object feature is a feature of a target object of a lost item;

[0223] The video segment determination module 702 is configured to determine, in response to the item search request, a second video segment in which the item is lost based on identification information of a first video segment captured by at least two cameras at different locations, wherein the first video segment includes a target object, the identification information includes a feature of the target object and an item tag associated with the feature of the target object, and the item tag indicates an item carried by the target object in the first video segment;

[0224] The result sending module 703 is used to send the object search result to the terminal, where the object search result includes the position information corresponding to the camera that captures the second video clip.

[0225] In a possible implementation, the video segment determination module 702 is configured to:

[0226] For every two first video segments that are captured at adjacent times and by cameras at different locations, if there is identification information of the first target video segment whose object tag is included in the two first video segments and identification information of the second target video segment that is not included in the two first video segments, at least one of the first target video segment and the second target video segment is determined as the second video segment;

[0227] The acquisition time of the first target video segment is earlier than the acquisition time of the second target video segment.

[0228] In a possible implementation, the object search request also carries a target object tag, where the target object tag indicates the lost object; the video segment determination module 702 is configured to:

[0229] If the target object tag is included in the identification information of the first target video segment and is not included in the identification information of the second target video segment, at least one of the first target video segment and the second target video segment is determined as the second video segment.

[0230] In a possible implementation, the device further includes:

[0231] The video segment filtering module is used to filter the other second video segments except the second video segment with the latest acquisition time when multiple second video segments are determined based on multiple first video segments.

[0232] In a possible implementation, the video segment determination module 702 is configured to:

[0233] For every two adjacent video frames in the first target video segment, if the identification information of the first video frame of the two video frames includes the object tag, and the identification information of the second video frame of the two video frames does not include the object tag, then the second video frame is determined to be the target video frame where the object is lost, the first target video segment is determined to be the second video segment, and the acquisition time of the first video frame is earlier than the acquisition time of the second video frame;

[0234] If the identification information of the last video frame in the first target video segment includes the object tag, the second target video segment is determined as the second video segment.

[0235] In a possible implementation, the item search request also carries a time period; the video segment determination module 702 is configured to:

[0236] Based on identification information of a first video segment captured by at least two cameras at different positions, a second video segment in which the item is lost and the capture time is within the time period is determined.

[0237] In a possible implementation, the device further includes:

[0238] A video segment extraction module, used to extract a first video segment containing a target object from videos captured by at least two cameras at different positions;

[0239] The identification information determination module is used to determine the identification information of the first video segment.

[0240] In a possible implementation, the identification information determination module is used to:

[0241] For each video frame in the first video segment, perform object recognition on the video frame to obtain an object region and object features contained in the object region;

[0242] Perform object recognition on the video frame to obtain the object area and the object labels corresponding to the objects contained in the object area;

[0243] When the object area is within the preset range of the object area, an association relationship is established between the object feature and the object tag.

[0244] In a possible implementation, the identification information determination module is used to:

[0245] The object recognition model is called to perform object recognition on the video frame to obtain the object area and the object labels corresponding to the objects contained in the object area.

[0246] In a possible implementation, the identification information determination module is used to:

[0247] Extracting a video frame from the first video segment every preset number of frames;

[0248] The extracted video frames are identified to obtain identification information.

[0249] In a possible implementation, the device further includes:

[0250] A video segment extraction module, used to extract video segments containing features of the same object from videos captured by at least two cameras at different locations;

[0251] An identification information determination module, used to determine identification information of a video clip;

[0252] a label adding module, for adding an item disappearance label corresponding to the item label in the identification information of at least one of the third video segment and the fourth video segment, for every two video segments that are collected at adjacent times and collected by cameras at different positions, if there is identification information of a third video segment whose item label is included in the two video segments and identification information of a fourth video segment that is not included in the two video segments, and the collection time of the third video segment is earlier than the collection time of the fourth video segment;

[0253] The video segment determination module 702 is used to search for a second target video segment whose identification information includes a missing object tag from the first video segments captured by at least two cameras at different positions, and determine that the second target video segment is the second video segment.

[0254] An embodiment of the present application provides a solution for locating the location of a lost item after it is lost. When searching for the lost item of a target object, for a first video clip captured by at least two cameras in different positions and containing the target object, since the identification information of the first video clip can indicate which items the target object carried in the first video clip, based on the identification information of the first video clip, the second video clip where the item is lost can be further located, thereby obtaining an item search result. The location indicated by the location information included in the item search result is likely to be the location where the item is lost, so that the target object who lost the item can find the item based on the location information, thereby achieving rapid determination of the location where the item is lost and improving the convenience of item search.

[0255] It should be noted that the device provided in the above embodiment, when implementing its functions, only uses the division of the above functional modules as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the server is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and their specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0256] Figure 8 is a schematic diagram of the structure of another object search device provided in an embodiment of the present application. Figure 8 , the device comprises:

[0257] The interface display module 801 is used to display the item search interface;

[0258] The feature collection module 802 is used to collect target object features corresponding to the target object based on the item search interface;

[0259] The request sending module 803 is used to send an item search request to the server, the item search request carries the target object feature, and the server is used to respond to the item search request, determine the second video segment where the item is lost based on the identification information of the first video segment captured by at least two cameras at different locations, and return the item search result, the item search result includes the location information of the camera that captured the second video segment;

[0260] A result receiving module 804 is used to receive the item search result and display the item search result;

[0261] The first video clip contains a target object, and the identification information includes features of the target object and an item tag associated with the features of the target object, and the item tag indicates an item carried by the target object in the first video clip.

[0262] In a possible implementation, the item search request further carries a target item tag, where the target item tag indicates the lost item, and the device further includes:

[0263] The label acquisition module is used to obtain the target item label entered in the item search interface.

[0264] In a possible implementation, the item search request also carries a time period, and the device further includes:

[0265] The time period acquisition module is used to obtain the time period entered in the item search interface.

[0266] The embodiment of the present application provides a solution for locating the location of a lost item after it is lost. When the target object needs to find the item, it can trigger the terminal to send an item search request to the server, so that the lost item can be found with the help of the server. The server can locate the second video clip of the lost item based on the target object characteristics corresponding to the target object of the lost item, so as to obtain the item search result. The location indicated by the location information included in the item search result is likely to be the location where the item is lost, so that the target object of the lost item can find the item based on the location information, thereby realizing the rapid determination of the location where the item is lost and improving the convenience of item search.

[0267] It should be noted that the device provided in the above embodiment, when implementing its functions, only uses the division of the above functional modules as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the terminal is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and their specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0268] Fig. 9 : is a structural diagram of a server 900 provided in an embodiment of the present application. The server 900 may have relatively large differences due to different configurations or performances, and may include a processor (Central Processing Units, CPU) 901 and a memory 902, wherein the memory 902 is used to store at least one program code, and the at least one program code is loaded and executed by the processor 901 to implement the item search method in the above embodiment. Of course, the server 900 may also have components such as a wired or wireless network interface, a keyboard, and an input and output interface for input and output. The server 900 may also include other components for implementing terminal functions, which will not be described in detail here.

[0269] Those skilled in the art will understand that Fig. 9The structure shown in the figure does not constitute a limitation on the server 900, and the server 900 may include more or less components than those shown in the figure, or combine some components, or adopt a different component arrangement.

[0270] An embodiment of the present application provides a terminal, which includes a processor and a memory. The memory stores at least one program code, and the at least one program code is loaded and executed by the processor to implement the object search method in the above embodiment.

[0271] Fig.10 1 is a schematic diagram of the structure of a terminal 1000 provided according to an embodiment of the present application. The terminal 1000 may be a portable mobile terminal, such as a smart phone, a tablet computer, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 player (Moving Picture Experts Group Audio Layer IV), a laptop computer or a desktop computer. The terminal 1000 may also be referred to as a user device, a portable terminal, a laptop terminal, a desktop terminal or other names.

[0272] Typically, the terminal 1000 includes a processor 1001 and a memory 1002 .

[0273] The processor 1001 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 1001 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 1001 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 1001 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 1001 may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning.

[0274] The memory 1002 may include one or more computer-readable storage media, which may be non-transitory. The memory 1002 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 1002 is used to store at least one computer program, which is executed by the processor 1001 to implement the object search method provided in the method embodiment of the present application.

[0275] In some embodiments, the terminal 1000 may also optionally include: a peripheral device interface 1003 and at least one peripheral device. The processor 1001, the memory 1002 and the peripheral device interface 1003 may be connected via a bus or a signal line. Each peripheral device may be connected to the peripheral device interface 1003 via a bus, a signal line or a circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit 1004, a display screen 1005, a camera assembly 1006, an audio circuit 1007, a positioning assembly 1008 and a power supply 1009.

[0276] The peripheral device interface 1003 may be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 1001 and the memory 1002. In some embodiments, the processor 1001, the memory 1002, and the peripheral device interface 1003 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 1001, the memory 1002, and the peripheral device interface 1003 may be implemented on a separate chip or circuit board, which is not limited in this embodiment.

[0277] The radio frequency circuit 1004 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit 1004 communicates with the communication network and other communication devices through electromagnetic signals. The radio frequency circuit 1004 converts the electrical signal into an electromagnetic signal for transmission, or converts the received electromagnetic signal into an electrical signal. Optionally, the radio frequency circuit 1004 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, etc. The radio frequency circuit 1004 can communicate with other terminals through at least one wireless communication protocol. The wireless communication protocol includes but is not limited to: the World Wide Web, a metropolitan area network, an intranet, various generations of mobile communication networks (2G, 3G, 4G and 5G), a wireless local area network and / or a WiFi (Wireless Fidelity) network. In some embodiments, the radio frequency circuit 1004 may also include circuits related to NFC (Near Field Communication), which is not limited in this application.

[0278] The display screen 1005 is used to display a UI (User Interface). The UI may include graphics, text, icons, videos, and any combination thereof. When the display screen 1005 is a touch display screen, the display screen 1005 also has the ability to collect touch signals on the surface or above the surface of the display screen 1005. The touch signal can be input to the processor 1001 as a control signal for processing. At this time, the display screen 1005 can also be used to provide virtual buttons and / or virtual keyboards, also known as soft buttons and / or soft keyboards. In some embodiments, the display screen 1005 can be one, arranged on the front panel of the terminal 1000; in other embodiments, the display screen 1005 can be at least two, respectively arranged on different surfaces of the terminal 1000 or in a folding design; in other embodiments, the display screen 1005 can be a flexible display screen, arranged on a curved surface or a folding surface of the terminal 1000. Even, the display screen 1005 can also be arranged as a non-rectangular irregular figure, that is, a special-shaped screen. The display screen 1005 can be made of materials such as LCD (Liquid Crystal Display) and OLED (Organic Light-Emitting Diode).

[0279] The camera assembly 1006 is used to capture images or videos. Optionally, the camera assembly 1006 includes a front camera and a rear camera. Typically, the front camera is arranged on the front panel of the terminal, and the rear camera is arranged on the back of the terminal. In some embodiments, there are at least two rear cameras, which are any one of a main camera, a depth of field camera, a wide-angle camera, and a telephoto camera, so as to realize the fusion of the main camera and the depth of field camera to realize the background blur function, the fusion of the main camera and the wide-angle camera to realize the panoramic shooting and VR (Virtual Reality) shooting function or other fusion shooting functions. In some embodiments, the camera assembly 1006 may also include a flash. The flash can be a monochrome temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cold light flash, which can be used for light compensation at different color temperatures.

[0280] The audio circuit 1007 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, and convert the sound waves into electrical signals and input them into the processor 1001 for processing, or input them into the radio frequency circuit 1004 to achieve voice communication. For the purpose of stereo acquisition or noise reduction, there may be multiple microphones, which are respectively arranged at different parts of the terminal 1000. The microphone may also be an array microphone or an omnidirectional acquisition microphone. The speaker is used to convert the electrical signal from the processor 1001 or the radio frequency circuit 1004 into sound waves. The speaker may be a traditional film speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can not only convert the electrical signal into sound waves audible to humans, but also convert the electrical signal into sound waves inaudible to humans for purposes such as ranging. In some embodiments, the audio circuit 1007 may also include a headphone jack.

[0281] The positioning component 1008 is used to locate the current geographical location of the terminal 1000 to implement navigation or LBS (Location Based Service). The positioning component 1008 can be a positioning component based on the US GPS (Global Positioning System), China's Beidou system, Russia's Grenas system, or the European Union's Galileo system.

[0282] The power supply 1009 is used to power various components in the terminal 1000. The power supply 1009 can be an alternating current, a direct current, a disposable battery, or a rechargeable battery. When the power supply 1009 includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery charged through a wired line, and a wireless rechargeable battery is a battery charged through a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0283] Those skilled in the art will understand that Fig.10 The structure shown in the figure does not constitute a limitation on the terminal 1000, and the terminal 1000 may include more or less components than those shown in the figure, or combine some components, or adopt a different component arrangement.

[0284] In an exemplary embodiment, a computer-readable storage medium is also provided, in which at least one program code is stored, and the at least one program code is loaded and executed by a processor to implement the object search method in the above embodiment. The computer-readable storage medium can be a memory. For example, the computer-readable storage medium can be a ROM (Read-Only Memory), a RAM (Random Access Memory), a CD-ROM (Compact Disc Read-Only Memory), a magnetic tape, a floppy disk, an optical data storage terminal, etc.

[0285] In an exemplary embodiment, a computer program product is also provided, which includes a computer program code, the computer program code is stored in a computer-readable storage medium, a processor reads the computer program code from the computer-readable storage medium, and the processor executes the computer program code to implement the object search method as in the above embodiment.

[0286] In some embodiments, the computer program involved in the embodiments of the present application may be deployed and executed on a computer device, or on multiple computer devices located in one location, or on multiple computer devices distributed in multiple locations and interconnected through a communication network, and multiple computer devices distributed in multiple locations and interconnected through a communication network may constitute a blockchain system. The computer device may be provided as a terminal or a server.

[0287] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0288] The above are only optional embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for searching for an item, characterized in that: The method comprises: An item search request sent by a receiving terminal is carried by the item search request, and the target object feature is a feature of a target object of the lost item; In response to the object search request, for each of at least two cameras at different locations, splicing video frames that recognize the target object and have adjacent acquisition times to obtain a first video segment that contains the target object and is acquired by the camera; obtaining the first video segment corresponding to each camera, and arranging the multiple first video segments according to the order of arrangement of the time periods; Determining identification information of the first video clip, the identification information of the first video clip including object features identified from video frames of the first video clip and item tags associated with the object features, the object features including target object features and other object features, the target item tags associated with the target object features indicating items carried by the target object in the first video clip; For every two first video clips captured by cameras at adjacent acquisition times and at different locations, if there is identification information of a first target video clip in which the target item tag is included in the two first video clips, and identification information of a second target video clip not included in the two first video clips, at least one of the first target video clip and the second target video clip is determined to be the second video clip where the item is lost, and the acquisition time of the first target video clip is earlier than the acquisition time of the second target video clip; Sending an object search result to the terminal, the object search result including location information corresponding to the camera that captures the second video clip.

2. The method according to claim 1, characterized in that The item search request also carries the target item tag, where the target item tag represents the lost item; the method further includes: If the target object tag is included in the identification information of the first target video segment and is not included in the identification information of the second target video segment, at least one of the first target video segment and the second target video segment is determined as the second video segment.

3. The method according to claim 1, characterized in that After determining at least one of the first target video segment and the second target video segment as the second video segment where the item is lost, and before sending the item search result to the terminal, the method further includes: When a plurality of second video segments are determined based on a plurality of first video segments, the second video segments other than the second video segment with the latest acquisition time are filtered.

4. The method according to claim 1, characterized in that: The step of determining at least one of the first target video segment and the second target video segment as a second video segment with lost quality includes: For every two adjacent video frames in the first target video segment, if the identification information of the first video frame of the two video frames includes the item tag, and the identification information of the second video frame of the two video frames does not include the item tag, then determining the second video frame as the target video frame where the item is lost, determining the first target video segment as the second video segment, and the acquisition time of the first video frame is earlier than the acquisition time of the second video frame; If the identification information of the last video frame in the first target video segment includes the object tag, the second target video segment is determined as the second video segment.

5. The method according to claim 1, characterized in that The item search request also carries a time period; the method further includes: Based on the identification information of the first video segment captured by at least two cameras in different positions, a second video segment in which the item is lost and the capture time is within the time period is determined.

6. The method according to claim 1, characterized in that The determining the identification information of the first video segment includes: For each video frame in the first video clip, performing object recognition on the video frame to obtain an object region and object features contained in the object region; Performing object recognition on the video frame to obtain an object region and an object label corresponding to the object contained in the object region; When the item area is within a preset range of the object area, an association relationship is established between the object feature and the item tag.

7. The method according to claim 6, characterized in that The performing object recognition on the video frame to obtain an object region and an object label corresponding to the object contained in the object region includes: An object recognition model is called to perform object recognition on the video frame to obtain the object region and object labels corresponding to the objects contained in the object region.

8. The method according to claim 1, characterized in that: The determining the identification information of the first video segment includes: Extracting a video frame from the first video segment every preset number of frames; The extracted video frame is identified to obtain the identification information.

9. The method according to claim 1, characterized in that: The method further comprises: Extracting video segments containing the same object from videos captured by at least two cameras at different locations; Determining identification information of the video clip; For every two video clips that are collected at adjacent times and by cameras at different positions, if there is identification information of a third video clip whose item tag is included in the two video clips, and identification information of a fourth video clip that is not included in the two video clips, then in the identification information of at least one of the third video clip and the fourth video clip, an item disappearance tag corresponding to the item tag is added, and the collection time of the third video clip is earlier than the collection time of the fourth video clip; The method further comprises: From the first video segments captured by at least two cameras at different positions, a second target video segment whose identification information includes the missing tag of the object is searched, and the second target video segment is determined to be the second video segment.

10. A method for searching for an item, characterized in that: The method comprises: Display the item search interface; Based on the item search interface, collecting target object features corresponding to the target object; Sending an item search request to a server, wherein the item search request carries the target object feature; The server is used to: for each camera of at least two cameras located in different positions, splice the video frames with adjacent acquisition time that identify the target object to obtain a first video segment collected by the camera and containing the target object; obtain the first video segment corresponding to each camera, and arrange the multiple first video segments according to the arrangement order of the time periods; determine the identification information of the first video segment, the identification information of the first video segment includes the object features identified from the video frames of the first video segment and the object tags associated with the object features, the object features include the target object features and other object features, and the target object tags associated with the target object features refer to The method further comprises: showing an item carried by the target object in the first video clip; for every two first video clips collected by cameras with adjacent collection times and at different positions, if there is identification information of the first target video clip whose target item tag is included in the two first video clips, and identification information of the second target video clip not included in the two first video clips, then determining at least one of the first target video clip and the second target video clip as the second video clip where the item is lost, and the collection time of the first target video clip is earlier than the collection time of the second target video clip; returning an item search result, wherein the item search result includes the location information corresponding to the camera that collected the second video clip; The item search result is received and displayed.

11. The method according to claim 10, characterized in that The item search request also carries a target item tag, where the target item tag indicates the lost item. The method further includes: The target item label input in the item search interface is obtained.

12. The method according to claim 10, characterized in that The item search request also carries a time period, and the method further includes: The time period entered in the item search interface is obtained.

13. An object search device, characterized in that: The device comprises: A request receiving module, configured to receive an item search request sent by a terminal, wherein the item search request carries a target object feature, and the target object feature is a feature of a target object of a lost item; a video segment determination module, configured to respond to the item search request, for each camera of at least two cameras at different locations, splice the video frames that identify the target object and have adjacent acquisition times, to obtain a first video segment that contains the target object acquired by the camera; obtain the first video segment corresponding to each camera, and arrange the multiple first video segments according to the arrangement order of the time periods; determine the identification information of the first video segment, the identification information of the first video segment including the object features identified from the video frames of the first video segment and the item tags associated with the object features, the object features including the target object features and other object features, the target item tags associated with the target object features indicating the items carried by the target object in the first video segment; for every two first video segments acquired by cameras with adjacent acquisition times and at different locations, if there is identification information of the first target video segment whose target item tag is included in the two first video segments, and identification information of the second target video segment that is not included in the two first video segments, then at least one of the first target video segment and the second target video segment is determined as the second video segment where the item is lost, and the acquisition time of the first target video segment is earlier than the acquisition time of the second target video segment; The result sending module is used to send the object search result to the terminal, and the object search result includes the position information corresponding to the camera that captures the second video clip.

14. An object search device, characterized in that: The device comprises: An interface display module, used to display an item search interface; A feature collection module, used for collecting target object features corresponding to the target object based on the item search interface; A request sending module, used to send an item search request to a server, wherein the item search request carries the target object feature; The server is used to: for each camera of at least two cameras located in different positions, splice the video frames with adjacent acquisition time that identify the target object to obtain a first video segment collected by the camera and containing the target object; obtain the first video segment corresponding to each camera, and arrange the multiple first video segments according to the arrangement order of the time periods; determine the identification information of the first video segment, the identification information of the first video segment includes the object features identified from the video frames of the first video segment and the object tags associated with the object features, the object features include the target object features and other object features, and the target object tags associated with the target object features refer to The method further comprises: showing an item carried by the target object in the first video clip; for every two first video clips collected by cameras with adjacent collection times and at different positions, if there is identification information of the first target video clip whose target item tag is included in the two first video clips, and identification information of the second target video clip not included in the two first video clips, then determining at least one of the first target video clip and the second target video clip as the second video clip where the item is lost, and the collection time of the first target video clip is earlier than the collection time of the second target video clip; returning an item search result, wherein the item search result includes the location information corresponding to the camera that collected the second video clip; The result receiving module is used to receive the item search result and display the item search result.

15. A server, characterized in that: The server includes a processor and a memory, wherein the memory stores at least one program code, and the at least one program code is loaded and executed by the processor to implement the item search method according to any one of claims 1 to 9.

16. A terminal, characterized in that: The terminal includes a processor and a memory, wherein at least one program code is stored in the memory, and the at least one program code is loaded and executed by the processor to implement the object search method according to any one of claims 10 to 12.

17. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one program code, and the at least one program code is loaded and executed by the processor to implement the object search method as described in any one of claims 1 to 9, or to implement the object search method as described in any one of claims 10 to 12.

Images