Tracking method and device for shooting teacher close-up picture, equipment and storage medium

Abstract

The invention discloses a tracking method and device for shooting teacher close-up pictures, equipment and a storage medium. Wherein. The method comprises the following steps: acquiring a current tracking state; when tracking target shooting is locked according to the face information, when the tracking state is changed from a locked tracking target to a tracking target loss state, executing a first strategy; the first strategy comprises the following steps: enabling the teacher close-up camera to keep a current shooting position for a first preset time length; in a first preset duration, in the teacher close-up picture, an auxiliary mode is used for detection, and the auxiliary mode is human head detection or human body detection; and when the object is detected by using the auxiliary mode, re-locking the object detected by using the auxiliary mode as the tracking target, and switching the teacher close-up camera to lock the tracking target according to the information detected by using the auxiliary mode for shooting.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and more specifically, to a tracking method, device, equipment and storage medium for shooting close-up images of teachers. Background Art

[0002] Artificial intelligence visual target automatic tracking shooting is widely used in various fields and scenarios, such as teacher tracking or student tracking shooting in remote teaching, tracking shooting of the speaker in a remote conference, etc. Taking the close-up shot of a teacher as an example, the teacher's facial information is usually matched with the facial information captured in the video to identify the target object in the video, and then the teacher's close-up camera is controlled to track and shoot the teacher according to the position of the target object.

[0003] However, in actual teaching scenarios, teachers do not always face the camera so that the camera can capture the teacher's face information. The teacher will write on the blackboard, etc. At this time, the teacher is facing away from the camera, and the camera cannot capture the teacher's face information. When the camera cannot capture the teacher's face, it defaults to losing the target object and automatically returns to the preset position to capture again. In this way, the close-up camera of the teacher will capture a strange picture, which makes the viewer uncomfortable. Summary of the invention

[0004] In order to solve the technical problem of improving the accuracy of shooting close-up images of teachers, the present invention provides a tracking method, device, equipment and storage medium for shooting close-up images of teachers. The technical solution adopted by the present invention is as follows.

[0005] In a first aspect, the present invention provides a tracking method for shooting a close-up picture of a teacher, wherein the tracking method for shooting a close-up picture of a teacher is used for tracking shooting within a pre-defined podium area, and the close-up picture of the teacher is a picture shot by a close-up camera of the teacher, comprising:

[0006] Get the current tracking status;

[0007] When the tracking target is locked and photographed according to the face information, when the tracking state changes from the locked tracking target to the lost tracking target, the first strategy is executed; wherein the base database face used for locking the tracking target according to the face information is the teacher's face;

[0008] The first strategy includes the steps of:

[0009] Let the teacher's close-up camera maintain the current shooting position for a first preset duration;

[0010] Within the first preset time period, in the teacher close-up picture, detection is performed using an auxiliary method, wherein the auxiliary method is head detection or human body detection;

[0011] When an object is detected using the auxiliary method, the object detected using the auxiliary method is re-locked as a tracking target, and the teacher close-up camera is switched to lock the tracking target for shooting based on the information detected by the auxiliary method.

[0012] In one embodiment, it further includes:

[0013] When the tracking target is locked and photographed according to the information detected in the auxiliary mode, when the tracking state is that there is a locked tracking target, executing the second strategy;

[0014] The second strategy comprises the steps of:

[0015] Perform face detection on the tracking target;

[0016] When a face is detected, the detected face is verified with the face in the database;

[0017] When the verification result is a match, the detected face is re-locked as the tracking target, and the teacher's close-up camera is switched to lock the tracking target based on the face information for shooting.

[0018] In one embodiment, when the tracking target is locked and photographed according to the information detected in the auxiliary manner, when the tracking state changes from having a locked tracking target to having a tracking target lost, a third strategy is executed;

[0019] The third strategy comprises the steps of:

[0020] Let the teacher's close-up camera maintain the current shooting position for a second preset time period;

[0021] During the second preset time, in the teacher's close-up screen, the test is carried out using auxiliary methods;

[0022] When an object is detected using the auxiliary method, the object detected using the auxiliary method is re-locked as a tracking target, and the teacher's close-up camera continues to lock the tracking target and shoot according to the information detected by the auxiliary method.

[0023] In one embodiment, the first strategy further comprises the steps of:

[0024] When no object is detected within the first preset time period, the teacher close-up camera returns to the preset position.

[0025] In one embodiment, the second strategy further comprises the steps of:

[0026] When the verification result is mismatch, the teacher close-up camera returns to the preset position.

[0027] In one embodiment, the third strategy comprises the steps of:

[0028] When no object is detected within the second preset time period, the teacher close-up camera returns to the preset position.

[0029] A tracking method for shooting a close-up picture of a teacher in a classroom, characterized in that the tracking method for shooting a close-up picture of a teacher in a classroom is used for tracking shooting of a podium area that has been pre-demarcated, and the close-up picture of the teacher is a picture shot by a close-up camera of the teacher, comprising:

[0030] When in the pre-defined podium area, a tracking shooting strategy for the podium area is used, and the tracking shooting strategy for the podium area is the tracking method for shooting a close-up picture of the teacher as described above;

[0031] When outside the pre-demarcated podium area, use the tracking shooting strategy of the non-podium area;

[0032] The tracking shooting strategy of the non-podium area includes the following steps:

[0033] The target is locked and tracked according to the face information and photographed; wherein the face in the base database used for locking and tracking the target according to the face information is the teacher's face;

[0034] Get the current tracking status;

[0035] When the tracking state changes from having a locked tracking target to having a lost tracking target, executing a fourth strategy;

[0036] The fourth strategy comprises the steps of:

[0037] Let the teacher's close-up camera maintain the current shooting position for a third preset time period;

[0038] Within the third preset time period, face detection is performed in the teacher's close-up image;

[0039] When a face is detected, the detected face is verified with the face in the database;

[0040] When the verification result is a match, the face object is re-locked as a tracking target.

[0041] In one embodiment, the fourth strategy further comprises the steps of:

[0042] When no face is detected or the verification result is mismatch, the teacher close-up camera returns to the preset position.

[0043] In a second aspect, the present invention provides a tracking device for shooting a close-up picture of a teacher, wherein the tracking device for shooting a close-up picture of a teacher is used for tracking shooting within a pre-defined podium area, and the close-up picture of the teacher is a picture shot by a close-up camera of the teacher, and comprises:

[0044] The acquisition module is used to obtain the current tracking status;

[0045] A processing module, used for executing a first strategy when the tracking state changes from a locked tracking target to a lost tracking target when the tracking target is locked according to the face information and photographed; wherein the base database face used for locking the tracking target according to the face information is the teacher's face;

[0046] The first strategy includes the steps of:

[0047] Let the teacher's close-up camera maintain the current shooting position for a first preset duration;

[0048] Within the first preset time period, in the teacher close-up picture, detection is performed using an auxiliary method, wherein the auxiliary method is head detection or human body detection;

[0049] When an object is detected using the auxiliary method, the object detected using the auxiliary method is re-locked as a tracking target, and the teacher close-up camera is switched to lock the tracking target for shooting based on the information detected by the auxiliary method.

[0050] A tracking device for shooting a close-up picture of a teacher in a classroom, wherein the tracking device for shooting a close-up picture of a teacher in a classroom is used for tracking shooting of a pre-demarcated podium area, wherein the close-up picture of the teacher is a picture shot by a close-up camera of the teacher, comprising: an execution module;

[0051] The execution module is used to use the tracking shooting strategy of the podium area when the podium area is pre-demarcated, and the tracking shooting strategy of the podium area is implemented by the tracking device for shooting the close-up picture of the teacher as described above; when the podium area is outside the pre-demarcated podium area, the tracking shooting strategy of the non-podium area is used, and the tracking shooting strategy of the non-podium area includes the steps of:

[0052] The target is locked and tracked according to the face information and photographed; wherein the face in the base database used for locking and tracking the target according to the face information is the teacher's face;

[0053] Get the current tracking status;

[0054] When the tracking state changes from having a locked tracking target to having a lost tracking target, executing a fourth strategy;

[0055] The fourth strategy comprises the steps of:

[0056] Let the teacher's close-up camera maintain the current shooting position for a third preset time period;

[0057] Within the third preset time period, face detection is performed in the teacher's close-up image;

[0058] When a face is detected, the detected face is verified with the face in the database;

[0059] When the verification result is a match, the face object is re-locked as a tracking target.

[0060] In a third aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the above-mentioned methods when executing the program.

[0061] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method of any of the above-mentioned embodiments.

[0062] In the present invention, according to the characteristics of the teaching scene, the retrieval conditions in the podium area are relaxed, and the retrieval conditions outside the podium area are maintained. In the retrieval conditions in the podium area, when it is impossible to track and shoot based on face information, an auxiliary head detection or body detection method is temporarily used to capture the target object, and then the tracking and shooting based on face information is switched back to the tracking and shooting based on face information. The whole process of the present invention takes into account the two problems of "following the right person" and "not losing", and achieves the purpose of optimizing the close-up tracking of teachers to "not lose". BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a schematic diagram of the overall flow of an implementation method of Example 1 of the present invention.

[0064] Figure 2 It is a schematic diagram of the overall process of the first strategy of the first embodiment of the present invention.

[0065] Figure 3 It is a schematic diagram of the overall process of the second strategy of the first embodiment of the present invention.

[0066] Figure 4 It is a schematic diagram of the overall process of the third strategy of the first embodiment of the present invention.

[0067] Figure 5 It is a schematic diagram of the overall process of an implementation method of Example 21 of the present invention.

[0068] Figure 6 It is a schematic diagram of the overall structure of an implementation method of Example 31 of the present invention. DETAILED DESCRIPTION

[0069] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0070] It should be noted that the terms "first\second\..." involved in the embodiments of the present invention are only used to distinguish similar objects, and do not represent a specific order for the objects. It is understandable that the specific order or sequence of "first\second\..." can be interchanged where permitted. It should be understood that the objects distinguished by "first\second\..." can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein.

[0071] Embodiment 1

[0072] See also Figure 1 , Figure 1 1 is a flow chart of a tracking method for shooting a close-up picture of a teacher provided in Embodiment 1 of the present invention, the method comprising step S110 and step S120. It should be noted that step S110 and step S120 are merely reference numerals for clearly explaining the embodiment and the accompanying drawings. Figure 1 The corresponding relationship does not represent a limitation on the order of the steps in this embodiment. The tracking method for shooting a close-up picture of a teacher is used for tracking shooting within a pre-defined podium area, and the close-up picture of the teacher is a picture shot by a close-up camera of the teacher.

[0073] Step S110, obtaining the current tracking status;

[0074] Step S120, when the tracking target is locked and photographed according to the face information, when the tracking state changes from the locked tracking target to the lost tracking target, the first strategy is executed; wherein the base database face used for locking the tracking target according to the face information is the teacher's face;

[0075] like Figure 2 As shown, the first strategy includes the steps: step S210, step S220 and step S230.

[0076] Step S210, allowing the teacher's close-up camera to maintain the current shooting position for a first preset time period;

[0077] Step S220, within the first preset time period, in the teacher close-up picture, using an auxiliary method to detect, the auxiliary method being head detection or human body detection;

[0078] Step S230, when an object is detected using the auxiliary method, the object detected using the auxiliary method is re-locked as a tracking target, and the teacher's close-up camera is switched to lock the tracking target for shooting based on the information detected by the auxiliary method.

[0079] After being started or activated, the teacher close-up camera uses the teacher's face as the base database face to capture and lock the target person. At this time, the locked target person is the teacher. The base database face refers to the face photo pre-entered in the system. The base database face is used to identify and track the person in the video. After locking the target person, the teacher close-up camera follows the movement of the target person, tracks and shoots the target person, and outputs a close-up picture of the teacher. Among them, the teacher close-up camera can use an AI smart camera or a pan-tilt camera that can be used to identify, track and shoot people.

[0080] In step S110, the tracking state of the teacher close-up camera is obtained. When the teacher close-up camera has locked the target person and can track and shoot the target person, it is in the state of having locked tracking target; when the teacher close-up camera has not locked the target person and cannot track and shoot the target person, it is in the state of losing tracking target.

[0081] In addition, the tracking method for taking close-up shots of teachers has two means of locking the tracking target. One is based on facial information, that is, the acquired target face is verified and compared with the base map face. When the comparison is consistent, tracking and shooting is carried out based on the acquired target face; the other is to use an auxiliary method for detection, which is head detection or body detection, and the head detected by head detection or the body detected by body detection is used for tracking and shooting.

[0082] In this method, the default is to use tracking and shooting based on facial information. When tracking and shooting based on facial information is used, when the teacher turns his head, facial information cannot be collected, and the tracking state will change from a locked tracking target to a tracking target lost state. At this time, the first strategy is triggered.

[0083] In the first strategy, the teacher close-up camera is first allowed to maintain the current shooting position for a period of time, which is a preset time, such as 3 seconds. The reason for maintaining the current shooting position is that although the teacher is lost, the teacher is likely to be in the picture at this position. If the current position is maintained, the teacher is likely to be still photographed.

[0084] During the period of maintaining the current shooting position, the target object is captured as quickly as possible. In actual teaching scenarios, the main reason for losing the teacher is that the teacher turns his head to write on the blackboard, so at this time, the object detected by head detection or human body detection is likely to be the teacher. Then lock this object and continue to detect the target object by head detection or human body detection for tracking shooting, which can also achieve the effect of tracking and shooting the teacher.

[0085] Of course, in an actual classroom scenario, there may be obstructions blocking the teacher's body, so head detection is preferably used. In addition, since there is usually only one teacher in the podium area, even if head tracking is used, without face detection, there is a high probability that the right person is followed, and the person followed is the teacher. Therefore, this method does not need to continuously capture the position of the head during face tracking. After the head is captured by head detection, the position is used to determine whether it is the same head.

[0086] In one implementation, the first strategy further includes the following steps: Step S240.

[0087] Step S240, when no object is detected within the first preset time period, the teacher close-up camera returns to the preset position.

[0088] If the target object still cannot be detected within the first preset time and after switching to another detection mode, namely the auxiliary mode, it is very likely that the teacher has left the current position of the screen. Then let the teacher's close-up camera return to the preset position to re-capture the tracking object.

[0089] In one embodiment, the tracking method for capturing a close-up image of a teacher further includes: step S130.

[0090] Step S130, when the tracking target is locked and photographed according to the information detected in the auxiliary mode, when the tracking state is that there is a locked tracking target, executing the second strategy;

[0091] like Figure 3 As shown, the second strategy includes the steps: step S310, step S320 and step S330.

[0092] Step S310, performing face detection on the tracking target;

[0093] Step S320, when a face is detected, verifying the detected face with the face in the database;

[0094] Step S330, when the verification result is a match, the face object is re-locked as a tracking target, and the teacher close-up camera is switched to lock the tracking target according to the face information for shooting.

[0095] Using auxiliary methods to detect and capture the target object may result in errors, and other non-teacher objects may be mistakenly locked as the target object. Therefore, it can only serve as a temporary method to track the target object. When face information can be used for tracking shooting, face information should still be used for tracking shooting. Therefore, after switching to using auxiliary methods to detect and capture the target object, the second strategy should be used to switch the tracking shooting mode back to the face information tracking shooting mode.

[0096] Steps S310 to S320 are the process of determining whether facial information can be used for tracking and shooting. When the verification result is a match, that is, when it is determined that the teacher's face can be recaptured, step S330 switches back to the facial information tracking and shooting mode.

[0097] In one embodiment, the second strategy further includes the following steps: Step S340.

[0098] Step S340, when the verification result is mismatch, the teacher close-up camera returns to the preset position.

[0099] If the verification result is mismatch, it means that the wrong target object has been found. In this case, the teacher can return the close-up camera to the preset position and re-capture the tracked object.

[0100] In one embodiment, when the tracking target is locked and photographed according to the information detected in the auxiliary manner, when the tracking state changes from having a locked tracking target to having a tracking target lost, a third strategy is executed;

[0101] like Figure 4 As shown, the third strategy includes the steps: step S410, step S420 and step S430.

[0102] Step S410, allowing the teacher's close-up camera to maintain the current shooting position for a second preset time period;

[0103] Step S420, performing detection in an auxiliary manner in the teacher close-up image within a second preset time period;

[0104] Step S430, when an object is detected using the auxiliary method, the object detected using the auxiliary method is re-locked as a tracking target, and the teacher's close-up camera continues to lock the tracking target and shoot according to the information detected by the auxiliary method.

[0105] After switching to tracking the target through head detection or human body detection, the tracking may also be lost. The processing process at this time is similar to the first strategy. First, let the teacher's close-up camera maintain the current shooting position for a period of time, that is, the second preset time, such as 3 seconds, and use this time to continue to use the auxiliary method to detect the target object. If the target object can be detected, then continue to lock the tracking target and shoot according to the information detected by the auxiliary method. Since the camera remains in the original position throughout the process, although the background is capturing the tracking target, for the viewer, it still feels like the teacher is being tracked and filmed.

[0106] In one implementation, the third strategy further includes the following steps: Step S440.

[0107] Step S440: When no object is detected within the second preset time period, the teacher close-up camera returns to the preset position.

[0108] If no object is detected within the second preset time period, it means that the teacher has most likely left the current position of the screen, so the teacher's close-up camera returns to the preset position to re-capture the tracked object.

[0109] In this method, according to the characteristics of the teaching scene, the retrieval conditions in the podium area are relaxed. When tracking and shooting based on facial information is not possible, auxiliary head detection or body detection methods are temporarily used to capture the target object, and then the tracking and shooting based on facial information is switched back to the right person when the opportunity arises. The whole process takes into account both the problems of "following the right person" and "not losing the person", thereby achieving the purpose of optimizing the close-up tracking of teachers and "not losing the person".

[0110] Embodiment 2

[0111] See also Figure 5 , Figure 5 1 is a flow chart of a tracking method for capturing a close-up image of a teacher in a classroom provided by Embodiment 2 of the present invention. The method includes step S510 and step S520. It should be noted that step S510 and step S520 are merely reference numerals for clearly explaining the embodiment and the appendix. Figure 5 The corresponding relationship does not represent a limitation on the order of the steps in this embodiment. The tracking method for shooting a close-up picture of a teacher in a classroom is used for tracking shooting of a pre-demarcated podium area, and the close-up picture of the teacher is the picture shot by the close-up camera of the teacher.

[0112] S510, when in the pre-defined podium area, using the tracking shooting strategy of the podium area, the tracking shooting strategy of the podium area is the tracking method for shooting a close-up picture of the teacher as described in the first embodiment;

[0113] S520, when outside the pre-defined podium area, using a tracking shooting strategy for the non-podium area;

[0114] The tracking shooting strategy for the non-podium area includes the following steps: step S610, step S620 and step S630.

[0115] Step S610, locking the tracking target according to the face information and shooting; wherein the face in the database used to lock the tracking target according to the face information is the teacher's face;

[0116] Step S620, obtaining the current tracking status;

[0117] Step S630, when the tracking state changes from having a locked tracking target to having a tracking target lost, executing a fourth strategy;

[0118] The fourth strategy includes the following steps: step S710, step S720, step S730 and step S740.

[0119] Step S710, allowing the teacher close-up camera to maintain the current shooting position for a third preset time period;

[0120] Step S720, performing face detection in the teacher close-up picture within the third preset time period;

[0121] Step S730, when a face is detected, verifying the detected face with the face in the database;

[0122] Step S740: When the verification result is a match, the face object is re-locked as a tracking target.

[0123] In the second embodiment, the area inside the podium and the area outside the podium are divided, and different tracking methods are used for different areas. In the podium area, the tracking strategy of the first embodiment is used, and outside the podium area, general tracking and shooting based on facial information is adopted, that is, as shown in the fourth strategy, when the target is lost, the face in the teacher's close-up picture is detected, and when a matching face is detected, the tracking target is locked again. The fourth strategy is similar to the previous strategy, and both let the teacher's close-up camera maintain the current shooting position for a period of time, and use this time to continue to recapture the target object.

[0124] The area is divided into the area inside the podium and the area outside the podium, and different tracking methods are used for different areas. This is because there is usually only the teacher in the area inside the podium. Even if head tracking is used without face detection, there is a high probability that the right person will be tracked, and the person being tracked is the teacher. There are usually students in the area outside the podium in the classroom. If head tracking is used in this area, it is very likely to result in tracking errors and students may be tracked as teachers. Therefore, face detection must be performed in the student area.

[0125] In one implementation, the fourth strategy further includes the following steps: Step S750.

[0126] Step S750, when no face is detected or the verification result is mismatch, the teacher close-up camera returns to the preset position.

[0127] If no face is detected within the third preset time period, it means that the teacher has most likely left the current position. If the verification result is a mismatch, it means that the wrong target object is found. In both cases, the teacher's close-up camera returns to the preset position to re-capture the tracked object.

[0128] In this method, according to the characteristics of the teaching scene, the retrieval conditions within the podium area are relaxed, while the retrieval conditions outside the podium area are maintained. The whole process takes into account both the issues of "following the right person" and "not losing the person", thus achieving the goal of optimizing the teacher's close-up tracking to "not lose the person".

[0129] Embodiment 3

[0130] Corresponding to the method of embodiment 1, as Figure 6 As shown, the present invention also provides a tracking device for shooting a close-up picture of a teacher, the tracking device for shooting a close-up picture of a teacher is used for tracking shooting within a pre-defined podium area, and the close-up picture of the teacher is a picture shot by a close-up camera of the teacher, including: an acquisition module 610, a processing module 620;

[0131] The acquisition module is used to obtain the current tracking status;

[0132] A processing module, used for executing a first strategy when the tracking state changes from a locked tracking target to a lost tracking target when the tracking target is locked according to the face information and photographed; wherein the base database face used for locking the tracking target according to the face information is the teacher's face;

[0133] The first strategy includes the steps of:

[0134] Let the teacher's close-up camera maintain the current shooting position for a first preset duration;

[0135] Within the first preset time period, in the teacher close-up picture, detection is performed using an auxiliary method, wherein the auxiliary method is head detection or human body detection;

[0136] When an object is detected using the auxiliary method, the object detected using the auxiliary method is re-locked as a tracking target, and the teacher close-up camera is switched to lock the tracking target for shooting based on the information detected by the auxiliary method.

[0137] In one embodiment, the processing module is further configured to execute a second strategy when the tracking state is that there is a locked tracking target when the tracking target is locked and photographed according to the information detected in the auxiliary manner;

[0138] The second strategy comprises the steps of:

[0139] Perform face detection on the tracking target;

[0140] When a face is detected, the detected face is verified with the face in the database;

[0141] When the verification result is a match, the detected face is re-locked as the tracking target, and the teacher's close-up camera is switched to lock the tracking target based on the face information for shooting.

[0142] In one embodiment, the processing module is further configured to execute a third strategy when the tracking state changes from a locked tracking target to a lost tracking target when the tracking target is locked and photographed according to the information detected in the auxiliary manner;

[0143] The third strategy comprises the steps of:

[0144] Let the teacher's close-up camera maintain the current shooting position for a second preset time period;

[0145] During the second preset time, in the teacher's close-up screen, the test is carried out using auxiliary methods;

[0146] When an object is detected using the auxiliary method, the object detected using the auxiliary method is re-locked as a tracking target, and the teacher's close-up camera continues to lock the tracking target and shoot according to the information detected by the auxiliary method.

[0147] In one embodiment, the first strategy further includes the step of: when no object is detected within a first preset time period, the teacher close-up camera returns to a preset position.

[0148] In one embodiment, the second strategy further includes the step of: when the verification result is mismatch, the teacher close-up camera returns to the preset position.

[0149] In one embodiment, the third strategy comprises the steps of:

[0150] When no object is detected within the second preset time period, the teacher close-up camera returns to the preset position.

[0151] Corresponding to the method of the second embodiment, the present invention further provides a tracking device for shooting a close-up picture of a teacher in a classroom, wherein the tracking device for shooting a close-up picture of a teacher in a classroom is used for tracking shooting of a pre-defined podium area, and the close-up picture of the teacher is a picture shot by a close-up camera of the teacher, comprising: an execution module;

[0152] The execution module is used to use the tracking shooting strategy of the podium area when the podium area is pre-demarcated, and the tracking shooting strategy of the podium area is implemented by the tracking device for shooting the close-up picture of the teacher as described above; when the podium area is outside the pre-demarcated podium area, the tracking shooting strategy of the non-podium area is used, and the tracking shooting strategy of the non-podium area includes the steps of:

[0153] The target is locked and tracked according to the face information and photographed; wherein the face in the base database used for locking and tracking the target according to the face information is the teacher's face;

[0154] Get the current tracking status;

[0155] When the tracking state changes from having a locked tracking target to having a lost tracking target, executing a fourth strategy;

[0156] The fourth strategy comprises the steps of:

[0157] Let the teacher's close-up camera maintain the current shooting position for a third preset time period;

[0158] Within the third preset time period, face detection is performed in the teacher's close-up image;

[0159] When a face is detected, the detected face is verified with the face in the database;

[0160] When the verification result is a match, the face object is re-locked as a tracking target.

[0161] In one embodiment, the fourth strategy comprises the steps of:

[0162] When no face is detected or the verification result is mismatch, the teacher close-up camera returns to the preset position.

[0163] In this device, according to the characteristics of the teaching scene, the retrieval conditions in the podium area are relaxed, and the retrieval conditions outside the podium area are maintained. In the retrieval conditions in the podium area, when it is impossible to track and shoot based on face information, auxiliary head detection or body detection methods are temporarily used to capture the target object, and then switch back to tracking and shooting based on face information when the opportunity arises. The entire process of this device takes into account the two issues of "following the right person" and "not losing", achieving the purpose of optimizing the close-up tracking of teachers and "not losing".

[0164] Embodiment 4

[0165] An embodiment of the present invention further provides a storage medium having computer instructions stored thereon, which, when executed by a processor, implements the tracking method for taking a close-up picture of a teacher in any of the above embodiments and / or the tracking method for taking a close-up picture of a teacher in a classroom.

[0166] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, random access memories (RAM), read-only memories (ROM), magnetic disks or optical disks, etc., various media that can store program codes.

[0167] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the relevant technology. The computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, terminal, or network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, RAM, ROM, magnetic disks or optical disks.

[0168] Corresponding to the above-mentioned computer storage medium, in one embodiment, a computer device is also provided, which includes a memory, an encoder, and a computer program stored in the memory and executable on the encoder, wherein when the encoder executes the program, it implements any one of the tracking methods for taking close-up pictures of teachers and / or the tracking method for taking close-up pictures of teachers in the classroom as described in the above-mentioned embodiments.

[0169] The above computer equipment, according to the characteristics of the teaching scene, relaxes the retrieval conditions in the podium area, maintains the retrieval conditions outside the podium area, and in the retrieval conditions in the podium area, when it is impossible to track and shoot based on facial information, first temporarily use auxiliary head detection or body detection methods to capture the target object, and then wait for an opportunity to switch back to tracking and shooting based on facial information. The entire process of the above computer equipment takes into account both the two issues of "following the right person" and "not losing", achieving the purpose of optimizing the close-up tracking of teachers to "not lose".

[0170] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0171] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A tracking method for shooting a close-up picture of a teacher, characterized in that: The tracking method for shooting a close-up picture of a teacher is used for tracking shooting within a pre-defined podium area, and the close-up picture of the teacher is a picture shot by a close-up camera of the teacher, including: Get the current tracking status; When the tracking target is locked and photographed according to the face information, when the tracking state changes from the locked tracking target to the lost tracking target, the first strategy is executed; wherein the base database face used for locking the tracking target according to the face information is the teacher's face; The first strategy includes the steps of: Let the teacher's close-up camera maintain the current shooting position for a first preset duration; Within the first preset time period, in the teacher close-up picture, detection is performed using an auxiliary method, wherein the auxiliary method is head detection or human body detection; When an object is detected using the auxiliary method, the object detected using the auxiliary method is re-locked as a tracking target, and the teacher close-up camera is switched to lock the tracking target for shooting based on the information detected by the auxiliary method.

2. The tracking method for taking close-up pictures of teachers according to claim 1, characterized in that: Also includes: When the tracking target is locked and photographed according to the information detected in the auxiliary mode, when the tracking state is that there is a locked tracking target, executing the second strategy; The second strategy comprises the steps of: Perform face detection on the tracking target; When a face is detected, the detected face is verified with the face in the database; When the verification result is a match, the detected face is re-locked as the tracking target, and the teacher's close-up camera is switched to lock the tracking target based on the face information for shooting.

3. The tracking method for taking close-up pictures of teachers according to claim 2, characterized in that: When the tracking target is locked and photographed according to the information detected in the auxiliary mode, when the tracking state changes from the locked tracking target to the tracking target being lost, executing the third strategy; The third strategy comprises the steps of: Let the teacher's close-up camera maintain the current shooting position for a second preset time period; During the second preset time, in the teacher's close-up screen, the test is carried out using auxiliary methods; When an object is detected using the auxiliary method, the object detected using the auxiliary method is re-locked as a tracking target, and the teacher's close-up camera continues to lock the tracking target and shoot according to the information detected by the auxiliary method.

4. The tracking method for taking close-up pictures of teachers according to any one of claims 1 to 3, characterized in that: The first strategy also includes the steps of: When no object is detected within the first preset time period, the teacher close-up camera returns to the preset position.

5. The tracking method for taking close-up pictures of teachers according to claim 2 or 3, characterized in that: The second strategy also includes the steps of: When the verification result is mismatch, the teacher close-up camera returns to the preset position.

6. The tracking method for taking close-up pictures of teachers according to claim 3, characterized in that: The third strategy comprises the steps of: When no object is detected within the second preset time period, the teacher close-up camera returns to the preset position.

7. A tracking method for taking close-up pictures of teachers in a classroom, characterized in that: The tracking method for shooting a close-up picture of a teacher in a classroom is used for tracking shooting of a pre-demarcated podium area, and the close-up picture of the teacher is a picture shot by a close-up camera of the teacher, including: When in the pre-defined podium area, a tracking shooting strategy for the podium area is used, wherein the tracking shooting strategy for the podium area is the tracking method for shooting a close-up image of a teacher as described in any one of claims 1 to 6; When outside the pre-demarcated podium area, use the tracking shooting strategy of the non-podium area; The tracking shooting strategy of the non-podium area includes the following steps: The target is locked and tracked according to the face information and photographed; wherein the face in the base database used for locking and tracking the target according to the face information is the teacher's face; Get the current tracking status; When the tracking state changes from having a locked tracking target to having a lost tracking target, executing a fourth strategy; The fourth strategy comprises the steps of: Let the teacher's close-up camera maintain the current shooting position for a third preset time period; Within the third preset time period, face detection is performed in the teacher's close-up image; When a face is detected, the detected face is verified with the face in the database; When the verification result is a match, the face object is re-locked as a tracking target.

8. The tracking method for taking close-up pictures of teachers in a classroom according to claim 7, characterized in that: The fourth strategy further comprises the steps of: When no face is detected or the verification result is mismatch, the teacher close-up camera returns to the preset position.

9. A tracking device for taking close-up pictures of teachers, characterized in that: The tracking device for shooting the teacher close-up picture is used for tracking shooting in a pre-defined podium area. The teacher close-up picture is a picture shot by the teacher close-up camera, including: The acquisition module is used to obtain the current tracking status; A processing module, used for executing a first strategy when the tracking state changes from a locked tracking target to a lost tracking target when the tracking target is locked according to the face information and photographed; wherein the base database face used for locking the tracking target according to the face information is the teacher's face; The first strategy includes the steps of: Let the teacher's close-up camera maintain the current shooting position for a first preset duration; Within the first preset time period, in the teacher close-up picture, detection is performed using an auxiliary method, wherein the auxiliary method is head detection or human body detection; When an object is detected using the auxiliary method, the object detected using the auxiliary method is re-locked as a tracking target, and the teacher close-up camera is switched to lock the tracking target for shooting based on the information detected by the auxiliary method.

10. A tracking device for taking close-up pictures of teachers in a classroom, characterized in that: The tracking device for shooting a close-up picture of a teacher in a classroom is used for tracking shooting of a pre-demarcated podium area, and the close-up picture of the teacher is a picture shot by a close-up camera of the teacher, and includes: an execution module; The execution module is used to use the tracking shooting strategy of the podium area when it is in the pre-defined podium area, and the tracking shooting strategy of the podium area is implemented by the tracking device for shooting close-up images of teachers according to claim 9; when it is outside the pre-defined podium area, the tracking shooting strategy of the non-podium area is used, and the tracking shooting strategy of the non-podium area includes the steps of: The target is locked and tracked according to the face information and photographed; wherein the face in the base database used for locking and tracking the target according to the face information is the teacher's face; Get the current tracking status; When the tracking state changes from having a locked tracking target to having a lost tracking target, executing a fourth strategy; The fourth strategy comprises the steps of: Let the teacher's close-up camera maintain the current shooting position for a third preset time period; Within the third preset time period, face detection is performed in the teacher's close-up image; When a face is detected, the detected face is verified with the face in the database; When the verification result is a match, the face object is re-locked as a tracking target.

11. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 8 is implemented.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images