Multi-target tracking face snapshot method and related equipment

Abstract

The invention provides a multi-target tracking face snapshot method and related equipment, and the method comprises the steps: obtaining tracking targets in a current frame image, and distributing an identification number for each tracking target; under the condition that the face identification number of the ith snapshot object is different from the identification number of any face tracking target and the human shape identification number of the ith snapshot object is the same as the identification number of any human shape tracking target, capturing and storing a first face image as a current face image of the ith snapshot object, and taking the identification number of the face tracking target associated with the first human shape tracking target as the face identification number of the ith snapshot object. The face tracking target is associated with the snapshot object based on the human shape tracking target, and the face tracking target does not need to newly create a snapshot object, so that the repeated snapshot rate is reduced.

Description

Technical Field

[0001] This invention relates to the field of image processing, and more specifically, to a method and related equipment for multi-target tracking and face capture. Background Technology

[0002] Currently, face detection and tracking technologies are widely used in security monitoring, intelligent access control, and other fields. Mainstream technologies include deep learning-based target detection algorithms (such as the YOLO series) and multi-target tracking algorithms (such as MOTR). These technologies can detect and track faces and human-shaped targets in videos in real time, but in complex scenes (such as occlusion, lighting changes, and target cross-movement), problems such as false detections, missed detections, and tracking loss still exist. Furthermore, existing technologies lack robust mechanisms for evaluating and filtering face image quality, resulting in inconsistent quality of captured face images, affecting the accuracy of subsequent recognition and analysis. Summary of the Invention

[0003] The purpose of this invention is to provide a face capture method and related equipment for multi-target tracking to improve the above-mentioned problems.

[0004] To achieve the above objectives, the technical solutions adopted in the embodiments of the present invention are as follows: In a first aspect, embodiments of the present invention provide a face capture method for multi-target tracking, the method comprising: The tracking targets in the current frame image are obtained, and each tracking target is assigned an identifier number, wherein the tracking targets include face tracking targets and human figure tracking targets; If the face identifier of the i-th captured object is different from the identifier of any face tracking target, and the human figure identifier of the i-th captured object is the same as the identifier of any human figure tracking target, then the first face image is captured and saved as the current face image of the i-th captured object. The identifier of the face tracking target associated with the first human figure tracking target is used as the face identifier of the i-th captured object. Here, the first face image is the face image corresponding to the face tracking target associated with the first human figure tracking target, and the first human figure tracking target is the human figure tracking target whose identifier is the same as the human figure identifier of the i-th captured object.

[0005] Secondly, embodiments of the present invention provide a face capture device for multi-target tracking, the device comprising: The first processing unit is used to acquire the tracking targets in the current frame image and assign an identifier to each tracking target, wherein the tracking targets include face tracking targets and human figure tracking targets; The second processing unit is configured to capture and save a first face image as the current face image of the i-th capture object when the face identifier of the i-th capture object is different from the identifier of any face tracking target, and the human figure identifier of the i-th capture object is the same as the identifier of any human figure tracking target. The unit also uses the identifier of the face tracking target associated with the first human figure tracking target as the face identifier of the i-th capture object. The first face image is the face image corresponding to the face tracking target associated with the first human figure tracking target, and the first human figure tracking target is the human figure tracking target whose identifier is the same as the human figure identifier of the i-th capture object.

[0006] Thirdly, embodiments of the present invention provide a storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method.

[0007] Fourthly, embodiments of the present invention provide an electronic device, the electronic device comprising: a processor and a memory, the memory being used to store one or more programs; when the one or more programs are executed by the processor, the above-described method is implemented.

[0008] Compared to existing technologies, the multi-target tracking face capture method and related equipment provided in this invention acquires tracking targets in the current frame image and assigns an identifier to each tracking target. The tracking targets include face tracking targets and human figure tracking targets. When the face identifier of the i-th capture object is different from the identifier of any face tracking target, and the human figure identifier of the i-th capture object is the same as the identifier of any human figure tracking target, a first face image is captured and saved as the current face image of the i-th capture object. The identifier of the face tracking target associated with the first human figure tracking target is used as the face identifier of the i-th capture object. The first face image is the face image corresponding to the face tracking target associated with the first human figure tracking target, and the first human figure tracking target is the human figure tracking target whose identifier is the same as the human figure identifier of the i-th capture object. By associating face tracking targets with capture objects based on human figure tracking targets, the face tracking target does not need to create a new capture object, thereby reducing the rate of repeated captures.

[0009] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0012] Figure 2 This is one of the flowcharts illustrating a face capture method for multi-target tracking provided in an embodiment of the present invention.

[0013] Figure 3 This is the second flowchart illustrating the face capture method for multi-target tracking provided in this embodiment of the invention.

[0014] Figure 4 This is the third flowchart illustrating the face capture method for multi-target tracking provided in this embodiment of the invention.

[0015] Figure 5 This is the fourth flowchart illustrating the multi-target tracking face capture method provided in this embodiment of the invention.

[0016] Figure 6 This is a schematic diagram of a multi-target tracking face capture device provided in an embodiment of the present invention.

[0017] In the diagram: 10-Processor; 11-Memory; 12-Bus; 13-Communication interface; 501-First processing unit; 502-Second processing unit. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0019] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0022] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of this invention is usually placed when in use. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention.

[0023] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set" and "connection" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0024] The following detailed description of some embodiments of the present invention is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0025] This invention provides an electronic device, which may be a server device, a computer device, or a mobile phone device, etc. Please refer to... Figure 1This is a schematic diagram of the structure of an electronic device. The electronic device includes a processor 10, a memory 11, and a bus 12. The processor 10 and the memory 11 are connected via the bus 12. The processor 10 is used to execute executable modules, such as computer programs, stored in the memory 11.

[0026] Processor 10 can be an integrated circuit chip with signal processing capabilities. During implementation, each step of the multi-target tracking face capture method can be completed through integrated logic circuits in the hardware or software instructions within processor 10. Processor 10 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0027] The memory 11 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage.

[0028] Bus 12 can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. Figure 1 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus 12 or one type of bus 12.

[0029] The memory 11 is used to store programs, such as programs corresponding to a multi-target tracking face capture device. The multi-target tracking face capture device includes at least one software functional module that can be stored in the memory 11 in the form of software or firmware, or embedded in the operating system (OS) of the electronic device. Upon receiving an execution instruction, the processor 10 executes the program to implement the multi-target tracking face capture method.

[0030] The electronic device provided in this embodiment of the invention may further include a communication interface 13. The communication interface 13 is connected to the processor 10 via a bus.

[0031] It should be understood that, Figure 1 The structure shown is only a partial schematic diagram of the electronic device; the electronic device may also include components that are larger than... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown. Figure 1 The components shown can be implemented using hardware, software, or a combination thereof.

[0032] The multi-target tracking face capture method provided in this embodiment of the invention can be applied to, but is not limited to, face capture. Figure 1 For the specific process of the electronic devices shown, please refer to [link / reference]. Figure 2 Multi-target tracking face capture methods, including S100 and S210, are described in detail below.

[0033] S100: Obtain the tracking targets in the current frame image and assign an identifier to each tracking target.

[0034] The tracking targets include face tracking targets and human figure tracking targets, and the identification number is a fixed and unique ID.

[0035] Optionally, in S100, the tracking targets in the current frame image are obtained, and an identifier is assigned to each tracking target, including S110 to S160, as follows.

[0036] S110, detect the initial target in the current frame image based on the target detection algorithm to obtain the initial target information group.

[0037] The object detection algorithm can be, but is not limited to, the YOLOv8 algorithm. The initial object information set includes the type of the initial object (face object or human-shaped object) and the coordinate set. The coordinate set of the initial object can be represented as (tx1, ty1, bx1, by1), where the coordinates of the top-left vertex of the minimum bounding rectangle of the initial object are (tx1, ty1), and the coordinates of the bottom-right vertex are (bx1, by1).

[0038] S120: Based on the initial target coordinate set, a target tracking algorithm is used to track the target in order to obtain the coordinate set corresponding to the tracked target.

[0039] The target tracking algorithm can be, but is not limited to, the MOTR algorithm. The set of coordinates corresponding to the tracked target can be represented as (tx2, ty2, bx2, by2).

[0040] S130, determine whether the tracking target matches any historical target in the identification information database. If they match, the identification number corresponding to the historical target is determined as the identification number of the tracking target. If they do not match, a new identification number is assigned to it, and the tracking target is added to the identification information database as a new historical target. The corresponding identification number is then updated in the identification information database.

[0041] The identifier database includes historical targets and their mapping relationships with corresponding identifiers. A tracked target matching a historical target can be defined as the similarity between the two being greater than a matching threshold.

[0042] S140, determine the overlap ratio between the tracking target and each initial target.

[0043] Optionally, the formula for calculating the overlap ratio is:

[0044]

[0045]

[0046] Where max is a function that finds the maximum value of the two, and min is a function that finds the minimum value of the two; The initial x-coordinate of the initial target coordinates. The x-coordinate of the initial target coordinates is the ending coordinate. Let y be the initial y-coordinate of the initial target coordinates. The y-coordinate is the end coordinate of the initial target coordinate; To track the starting x-coordinate of the target coordinates, To track the target coordinates, the x-coordinate ends. To track the starting y-coordinate of the target coordinates, To track the target coordinates, the y-coordinate ends. This indicates the overlap ratio. iou is the calculated ratio threshold, for example, 96.

[0047] S150, the initial target whose overlap with the tracked target is greater than the percentage threshold is used as the initial target for matching the tracked target; S160, determine the type of the target to be tracked based on the type of the initial target to be matched.

[0048] The types of targets to be tracked are face tracking targets or human figure tracking targets.

[0049] It should be understood that the tracking targets generated by the target tracking algorithm are generated in batches. The algorithm cannot identify their type, so it is necessary to calculate the overlap ratio with the initial target to determine the type of the tracking target.

[0050] In one alternative implementation, face tracking targets are stored in a face structure list, and human figure tracking targets are stored in a human figure structure list.

[0051] S210, if the face identifier of the i-th captured object is different from the identifier of any face tracking target, and the human figure identifier of the i-th captured object is the same as the identifier of any human figure tracking target, capture and save the first face image as the current face image of the i-th captured object, and use the identifier of the face tracking target associated with the first human figure tracking target as the face identifier of the i-th captured object.

[0052] At this point, the old face identifiers of the i-th captured object and their corresponding historical targets can be deleted from the identifier information database.

[0053] This refers to the i-th captured object in the capture list, where 1 ≤ i ≤ N, and N is the total number of captured objects in the capture list. The first face image is the face image corresponding to the face tracking target associated with the first human figure tracking target, and the first human figure tracking target is the human figure tracking target whose identifier is the same as the human figure identifier of the i-th captured object.

[0054] Optionally, using the first face image as the current face image of the i-th capture object can mean saving the first face image to the face image column corresponding to the i-th capture object in the capture list. In this embodiment of the invention, association means that the face tracking target is within the human shape tracking target.

[0055] Because faces are much smaller than human figures, they are difficult to track. Therefore, associating faces with human figures can greatly improve facial recognition. At the same time, when a person turns around, the face is lost. When the human figure turns around again, the relationship between the face and human figures can be re-established based on the human figure identifier. Based on these characteristics, the capture efficiency can be further improved on the basis of the tracking algorithm, the capture repetition rate can be reduced, storage space can be greatly saved, and the face image library can be continuously updated. At this time, the captured face image is the clearest image, which is convenient for subsequent processing. For example, if a person turns their back to the monitor after a long period of time, exceeding the time limit of current embedded tracking algorithms, the image may need to be re-captured. This solution completely solves this problem. Alternatively, if the target is too far away, the face is too small, or the face is too blurry to be detected, the tracking algorithm may fail to recognize the face as a new target. Therefore, a new identifier is assigned to the face, and face association can be achieved through this identifier, thereby reducing the rate of repeated captures.

[0056] Optionally, there may be situations where the face identifier of the i-th captured object is the same as the identifier of any face tracking target, and the human figure identifier of the i-th captured object is different from the identifier of any human figure tracking target. For example, the target is too close to the camera, resulting in no human figure being detected. For example, when there is only one human face in the whole picture, there may be situations where the human figure does not exist or the human figure is obscured by other objects, leaving only the face visible.

[0057] Please refer to Figure 3 When the face identifier of the i-th captured object is the same as the identifier of any face tracking target, and the human figure identifier of the i-th captured object is different from the identifier of any human figure tracking target, the face capture method for multi-target tracking also includes: S221, S222 and S223, which are described in detail below.

[0058] S221, Capture the second face image.

[0059] The second face image is the face image corresponding to the second face tracking target, and the second face tracking target is the face tracking target whose identifier is the same as the face identifier of the i-th captured object.

[0060] S222, when the second face image is complete and the clarity of the second face image is higher than the clarity of the current face image of the i-th captured object, the second face image is saved as the new current face image of the i-th captured object.

[0061] It should be understood that face images with lower resolution are deleted. At this time, the resolution value of the new current face image can be saved simultaneously to reduce the amount of calculation in the subsequent comparison process. At the same time, the historical targets corresponding to the face identifier of the i-th captured object in the identifier information database can also be updated.

[0062] Optionally, a Laplace filter can be used to obtain the sharpness value.

[0063] S223, use the identifier of the second humanoid tracking target as the humanoid identifier of the i-th captured object.

[0064] Optionally, the old human identification numbers and their corresponding historical targets preceding the i-th captured object are deleted from the identification information database.

[0065] Among them, the second human figure tracking target is the human figure tracking target associated with the second face tracking target.

[0066] Optionally, if the face identifier of the i-th captured object is the same as the identifier of any face tracking target, and the human figure identifier of the i-th captured object is the same as the identifier of any human figure tracking target, the face capture method for multi-target tracking further includes: S221 and S222, as follows.

[0067] S221, capture the second face image, wherein the second face image is the face image corresponding to the second face tracking target, and the second face tracking target is the face tracking target whose identifier is the same as the face identifier of the i-th captured object.

[0068] S222, when the second face image is complete and the clarity of the second face image is higher than the clarity of the current face image of the i-th captured object, the second face image is saved as the new current face image of the i-th captured object.

[0069] Optionally, "second face image complete" means that the coordinate region of the second face image does not coincide with the boundary of the current frame image.

[0070] Please refer to Figure 4 When the face identifier of the i-th captured object is different from the identifier of any face tracking target, and the human figure identifier of the i-th captured object is different from the identifier of any human figure tracking target, the face capture method for multi-target tracking also includes: S310 to S340, which are described in detail below.

[0071] S310, increment the lost frame count of the i-th captured object by 1.

[0072] S320: If the number of lost frames for any captured object exceeds a threshold, the lost object is removed from the capture list.

[0073] Among them, the lost object is the captured object whose frame count exceeds the threshold. The threshold can be, but is not limited to, 750.

[0074] S330, delete the face identifier of the lost object and its corresponding historical target from the identifier information database, and delete the human figure identifier of the lost object and its corresponding historical target from the identifier information database.

[0075] The identification information database includes historical targets and their mapping relationship with the corresponding identification numbers.

[0076] S340, if the face identifier of the i-th captured object is the same as the identifier of any face tracking target, or if the human figure identifier of the i-th captured object is the same as the identifier of any human figure tracking target, then reset the lost frame count of the i-th captured object to zero.

[0077] Please refer to Figure 5 In an optional implementation, the multi-target tracking face capture method further includes S410 and S420, which are described in detail below.

[0078] S410, when a third human figure is being tracked, a new capture object is created in the capture list. The human figure identifier of the new capture object is the identifier of the third human figure being tracked, and the face identifier of the new capture object is the identifier of the face tracking target associated with the third human figure being tracked. The current face image corresponding to the new capture object is captured (if there is no associated face tracking target identifier, the face identifier of the new capture object is empty, and the corresponding current face image is empty. It should be noted that the frame loss count of the new capture object is 0).

[0079] Among them, the human identification number of the third human tracking target is different from that of any captured object, and there is no associated identification number of the face tracking target, or the identification number of the associated face tracking target is different from that of any captured object.

[0080] S420, when a third face tracking target exists, a new capture object is created in the capture list. The face identifier of the new capture object is the identifier of the third face tracking target, and the human figure identifier of the new capture object is the identifier of the human figure tracking target associated with the third human figure tracking target. The current face image corresponding to the new capture object is captured (if there is no associated human figure tracking target identifier, the human figure identifier of the new capture object is empty. It should be noted that the frame loss count of the new capture object is 0).

[0081] Among them, the face identification number of the third face tracking target is different from that of any captured object, and there is no associated human tracking target identification number, or the identification number of the associated human tracking target is different from that of any captured object.

[0082] Please see Figure 6 , Figure 6 The present invention provides a multi-target tracking face capture device, which is optionally applied to the electronic device described above.

[0083] The multi-target tracking face capture device includes: a first processing unit 501 and a second processing unit 502.

[0084] The first processing unit 501 is used to acquire the tracking targets in the current frame image and assign an identifier to each tracking target, wherein the tracking targets include face tracking targets and human figure tracking targets; The second processing unit 502 is configured to, when the face identifier of the i-th captured object is different from the identifier of any face tracking target, and the human figure identifier of the i-th captured object is the same as the identifier of any human figure tracking target, capture and save the first face image as the current face image of the i-th captured object, and use the identifier of the face tracking target associated with the first human figure tracking target as the face identifier of the i-th captured object. Here, the first face image is the face image corresponding to the face tracking target associated with the first human figure tracking target, and the first human figure tracking target is the human figure tracking target whose identifier is the same as the human figure identifier of the i-th captured object.

[0085] Optionally, the first processing unit 501 may execute the above-described S100, and the second processing unit 502 may execute other steps in the above-described method embodiments.

[0086] It should be noted that the multi-target tracking face capture device provided in this embodiment can execute the method flow shown in the above-described method flow embodiment to achieve the corresponding technical effects. For the sake of brevity, any parts not mentioned in this embodiment can be referred to the corresponding content in the above-described embodiments.

[0087] This invention also provides a storage medium storing computer instructions and programs that, when read and executed, perform the multi-target tracking face capture method described above. The storage medium may include memory, flash memory, registers, or a combination thereof.

[0088] The following provides an electronic device, which may be a server device, a computer device, or a mobile phone device, etc. This electronic device, for example... Figure 1 As shown, the above-described face capture method with multi-target tracking can be implemented. Specifically, the electronic device includes: a processor 10, a memory 11, and a bus 12. The processor 10 may be a CPU. The memory 11 is used to store one or more programs, which, when executed by the processor 10, perform the face capture method with multi-target tracking described in the above embodiment.

[0089] In summary, the multi-target tracking face capture method and related device provided by this invention acquires tracking targets in the current frame image and assigns an identifier to each tracking target. The tracking targets include face tracking targets and human figure tracking targets. When the face identifier of the i-th capture object is different from the identifier of any face tracking target, and the human figure identifier of the i-th capture object is the same as the identifier of any human figure tracking target, a first face image is captured and saved as the current face image of the i-th capture object. The identifier of the face tracking target associated with the first human figure tracking target is used as the face identifier of the i-th capture object. The first face image is the face image corresponding to the face tracking target associated with the first human figure tracking target, and the first human figure tracking target is the human figure tracking target whose identifier is the same as the human figure identifier of the i-th capture object. By associating face tracking targets with capture objects based on human figure tracking targets, the face tracking target does not need to create a new capture object, thereby reducing the rate of repeated captures.

[0090] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0091] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A face snapshot method for multi-target tracking, characterized in that, The method comprises: acquiring tracking targets in a current frame image, and assigning an identification number to each tracking target, wherein the tracking targets comprise a face tracking target and a human shape tracking target; in a case where a face identification number of the ith snapshot object is different from identification numbers of any face tracking target, and a human shape identification number of the ith snapshot object is the same as an identification number of any human shape tracking target, capturing and saving a first face image as a current face image of the ith snapshot object, and taking an identification number of a face tracking target associated with a first human shape tracking target as the face identification number of the ith snapshot object, wherein the first face image is a face image corresponding to the face tracking target associated with the first human shape tracking target, and the first human shape tracking target is a human shape tracking target with the same identification number as the human shape identification number of the ith snapshot object.

2. The multi-target tracking face snapshot method of claim 1, wherein, in a case where the face identification number of the ith snapshot object is the same as an identification number of any face tracking target, and the human shape identification number of the ith snapshot object is different from identification numbers of any human shape tracking target, the method further comprises: capturing a second face image, wherein the second face image is a face image corresponding to a second face tracking target, and the second face tracking target is a face tracking target with the same identification number as the face identification number of the ith snapshot object; in a case where the second face image is complete, and a definition of the second face image is higher than a definition of a current face image of the ith snapshot object, saving the second face image as a new current face image of the ith snapshot object; taking an identification number of a second human shape tracking target as the human shape identification number of the ith snapshot object, wherein the second human shape tracking target is a human shape tracking target associated with the second face tracking target.

3. The multi-target tracking face snapping method of claim 1, wherein, in a case where the face identification number of the ith snapshot object is the same as an identification number of any face tracking target, and the human shape identification number of the ith snapshot object is the same as an identification number of any human shape tracking target, the method further comprises: capturing a second face image, wherein the second face image is a face image corresponding to a second face tracking target, and the second face tracking target is a face tracking target with the same identification number as the face identification number of the ith snapshot object; in a case where the second face image is complete, and a definition of the second face image is higher than a definition of a current face image of the ith snapshot object, saving the second face image as a new current face image of the ith snapshot object.

4. The multi-target tracking face snapshot method according to claim 2 or 3, wherein, the second face image being complete means that a coordinate region of the second face image does not coincide with a boundary of the current frame image.

5. The multi-target tracking face snapping method of claim 1, wherein, in a case where the face identification number of the ith snapshot object is different from identification numbers of any face tracking target, and the human shape identification number of the ith snapshot object is different from identification numbers of any human shape tracking target, the method further comprises: increasing a missing frame number count of the ith snapshot object by 1; in a case where the face identification number of the ith snapshot object is the same as an identification number of any face tracking target, or the human shape identification number of the ith snapshot object is the same as an identification number of any human shape tracking target, the method further comprises: clearing the missing frame number count of the ith snapshot object. In any case where the missing frame count of a snapped object exceeds a threshold, the method further comprises: deleting a missing object from the snapshot list, wherein the missing object is a snapped object whose missing frame count exceeds a threshold; deleting the face identifier number of the missing object and its corresponding historical target from the identification information library, and deleting the human form identifier number of the missing object and its corresponding historical target from the identification information library, the identification information library including a mapping relationship between historical targets and corresponding identifiers.

6. The multi-target tracking face snapping method of claim 1, wherein, The method further comprises: when there is a third human form tracking target, creating a new snapped object in the snapshot list, the human form identifier number of the new snapped object being the identifier number of the third human form tracking target, the face identifier number of the new snapped object being the identifier number of the human face tracking target associated with the third human form tracking target, and capturing the current face image corresponding to the new snapped object; wherein the third human form tracking target is different from the human form identifier number of any snapped object, and there is no associated human face tracking target identifier number, or the associated human form tracking target identifier number is different from the human form identifier number of any snapped object; when there is a third human face tracking target, creating a new snapped object in the snapshot list, the face identifier number of the new snapped object being the identifier number of the third human face tracking target, the human form identifier number of the new snapped object being the identifier number of the human form tracking target associated with the third human face tracking target, and capturing the current face image corresponding to the new snapped object; wherein the third human face tracking target is different from the face identifier number of any snapped object, and there is no associated human form tracking target identifier number, or the associated human form tracking target identifier number is different from the human form identifier number of any snapped object.

7. The multi-target tracking face snapping method of claim 1, wherein, The method further comprises: detecting initial targets in the current frame image based on a target detection algorithm to obtain an initial target information set, the initial target information set including the type and coordinate set of the initial target; tracking based on a target tracking algorithm based on the coordinate set of the initial target to obtain the coordinate set corresponding to the tracking target; determining whether the tracking target matches any historical target in the identification information library, if it matches, determining the identifier number corresponding to the historical target as the identifier number of the tracking target, if it does not match, assigning a new identifier number to it, and adding the tracking target as a new historical target to the identification information library, the identification information library including a mapping relationship between historical targets and corresponding identifiers; determining the overlap ratio of the tracking target with each initial target; taking the initial target with an overlap ratio with the tracking target greater than a ratio threshold as a matching initial target of the tracking target; determining the type of the tracking target according to the type of the matching initial target.

8. A multi-target tracking face snapping device, characterized in that, The device comprises: a first processing unit for obtaining tracking targets in a current frame image and assigning an identifier number to each tracking target, wherein the tracking targets include human face tracking targets and human form tracking targets; The second processing unit is configured to, in a case where the face identifier of the ith snapshot object is different from the identifier of any face tracking target, and the human body identifier of the ith snapshot object is the same as the identifier of any human body tracking target, capture and save the first face image as the current face image of the ith snapshot object, and set the identifier of the face tracking target associated with the first human body tracking target as the face identifier of the ith snapshot object, wherein the first face image is a face image corresponding to the face tracking target associated with the first human body tracking target, and the first human body tracking target is a human body tracking target having the same identifier as the human body identifier of the ith snapshot object.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program, when executed by a processor, implements the method of any one of claims 1-7.

10. An electronic device, comprising: comprising: a processor and a memory for storing one or more programs; when the one or more programs are executed by the processor, the method of any one of claims 1-7 is implemented.

Images