Teaching evaluation information generation method, device and equipment and readable storage medium

Abstract

According to the teaching evaluation information generation method and device, the equipment and the readable storage medium provided by the embodiment of the invention, the electronic equipment pulls the student panoramic stream from the student camera, generates the first thermodynamic diagram according to the student panoramic stream, and generates the teaching evaluation information of the target teacher according to the first thermodynamic diagram. Wherein the first thermodynamic diagram is used for indicating the moving track of the target teacher in the student area, the classroom where the target teacher is located comprises the teacher area and the student area, the student area is the visual field range of the student camera, and the teaching evaluation information is used for indicating the interaction condition between the target teacher and the students. According to the scheme, the real-time first thermodynamic diagram used for reflecting the moving track of the target teacher in the student area is generated, and the teaching evaluation information of the target teacher is generated according to the first thermodynamic diagram, so that teaching evaluation is performed from the perspective of interaction, and the purpose of improving the comprehensiveness of teaching evaluation results is achieved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of intelligent education technology, and in particular to a teaching evaluation information generation method, device, equipment and readable storage medium. Background Art

[0002] With the development of technology, more and more classrooms are equipped with recording and broadcasting systems. At the same time, in order to continuously improve the quality of teaching, it is necessary to evaluate the teaching level of teachers.

[0003] The current recording and broadcasting solution has an automatic broadcasting guidance function. Based on the automatic broadcasting guidance function, the recording and broadcasting system tracks the teacher in the video streams captured by the teacher camera and the student camera through a tracking algorithm. The teacher camera can be a camera used to record the teacher's teaching activities, which can usually be installed in a position where the teacher's activity area can be captured. The student camera refers to a camera that can be used to record teacher-student interaction activities, student classroom activities and other behaviors, which can usually be installed in a position where the student's activity area can be captured. When the interaction between the teacher and the student is tracked, the teacher's image is extracted separately as a close-up image of the teacher, and the teacher's behavior is analyzed based on the close-up image to achieve teaching evaluation.

[0004] However, the above teaching evaluation methods cannot directly reflect the classroom interaction situation, and the teaching evaluation results lack comprehensiveness. Summary of the invention

[0005] The embodiments of the present application provide a method, device, equipment and readable storage medium for generating teaching evaluation information. By generating a real-time trajectory heat map for reflecting the teacher's trajectory, teaching evaluation is performed from an interactive perspective based on the trajectory heat map, thereby achieving the purpose of improving the comprehensiveness of the teaching evaluation results.

[0006] In a first aspect, an embodiment of the present application provides a method for generating teaching evaluation information, which is applied to an electronic device, and the method includes:

[0007] Pull the student panoramic stream from the student camera;

[0008] Generate a first heat map according to the student panoramic flow, wherein the first heat map is used to indicate a movement trajectory of a target teacher in a student area, wherein the classroom where the target teacher is located includes a teacher area and the student area, and the student area is a field of view of the student camera;

[0009] The teaching evaluation information of the target teacher is generated according to the first heat map.

[0010] In a second aspect, an embodiment of the present application provides a teaching evaluation information generating device, comprising:

[0011] A pulling module is used to pull the student panoramic stream from the student camera;

[0012] A processing module, configured to generate a first heat map according to the student panoramic flow, wherein the first heat map is used to indicate a movement trajectory of a target teacher in a student area, wherein a classroom where the target teacher is located includes a teacher area and the student area, and the student area is a field of view of the student camera;

[0013] A generation module is used to generate teaching evaluation information of the target teacher based on the first heat map.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the electronic device implements the method described in the first aspect or various possible implementation methods in the first aspect.

[0015] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer instructions are stored. When the computer instructions are executed by a processor, they are used to implement the method described in the first aspect or various possible implementation methods of the first aspect.

[0016] In a fifth aspect, an embodiment of the present application provides a computer program product comprising a computing program, and when the computer program is executed by a processor, the method described in the first aspect or various possible implementation methods in the first aspect is implemented.

[0017] The teaching evaluation information generation method, device, equipment and readable storage medium provided in the embodiments of the present application, the electronic device pulls the student panoramic stream from the student camera, generates a first heat map based on the student panoramic stream, and generates the teaching evaluation information of the target teacher based on the first heat map. Among them, the first heat map is used to indicate the movement trajectory of the target teacher in the student area. The classroom where the target teacher is located includes a teacher area and a student area. The student area is the field of view of the student camera, and the teaching evaluation information is used to indicate the interaction between the target teacher and the student. With this scheme, by generating a real-time first heat map for reflecting the movement trajectory of the target teacher in the student area, the teaching evaluation information of the target teacher is generated based on the first heat map, so as to conduct teaching evaluation from the perspective of interaction, and achieve the purpose of improving the comprehensiveness of the teaching evaluation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1A It is a schematic diagram of an implementation environment of the teaching evaluation information generation method provided in the embodiment of the present application;

[0020] Figure 1B It is another implementation environment diagram of the teaching evaluation information generation method provided in the embodiment of the present application;

[0021] Figure 2 is a flow chart of a method for generating teaching evaluation information provided by an embodiment of the present application;

[0022] Figure 3 is another flow chart of the teaching evaluation information generation method provided by an embodiment of the present application;

[0023] Figure 4 It is the target heat map in the teaching evaluation information generation method provided in the embodiment of the present application;

[0024] Figure 5 It is a process diagram of the teaching evaluation information generation method provided in an embodiment of the present application;

[0025] Figure 6 A schematic diagram of a teaching evaluation information generating device provided in an embodiment of the present application;

[0026] Figure 7 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0027] The recording and broadcasting system is a system that uses a camera to record the scenes of teaching, meetings, performances, etc. for playback. More and more colleges, universities, and primary and secondary school classrooms are deploying recording and broadcasting systems. The recording and broadcasting system has an automatic broadcasting function: it uses a tracking algorithm to track the teacher in the video stream pulled by the teacher's camera and the student's camera. When the interaction between the teacher and the student is detected, the recording and broadcasting system generates a close-up picture of the teacher based on the video stream pulled from the teacher's camera or the student's camera.

[0028] Interactive behavior is an important indicator of teacher teaching evaluation. However, the above-mentioned close-up images of teachers cannot reflect the teacher's behavioral trajectory in the classroom, and cannot directly reflect the teacher's real-time interaction, resulting in the inability to evaluate teaching from an interactive perspective, which in turn leads to the evaluation dimension of the teaching evaluation results being not comprehensive enough.

[0029] Based on this, the embodiments of the present application provide a teaching evaluation information generation method, device, equipment and readable storage medium. Since the trajectory heat map can reflect whether the target teacher has been to the student area, the length of stay and range in the student area, etc., the electronic device performs teaching evaluation from an interactive perspective based on the trajectory heat map, thereby achieving the purpose of improving the comprehensiveness of the teaching evaluation results.

[0030] Figure 1A This is a schematic diagram of an implementation environment of the teaching evaluation information generation method provided in the embodiment of the present application. Figure 1A , a student camera 11 and a recording and broadcasting host 13 are set up in the classroom. The student camera 11 establishes a network connection with the recording and broadcasting host 13. The student camera 11 is located above the podium and is used to shoot the student area 15. The area above the dotted line in the figure including the podium is the teacher area 14, and the area below the dotted line is the student area 15.

[0031] During the lecture, the recording and broadcasting host 13 pulls the student panoramic stream from the student camera 11, and generates a first heat map in real time based on the student panoramic stream. The first heat map is used to indicate the movement trajectory of the target teacher in the student area. Based on the first heat map, teaching evaluation can be performed from an interactive perspective.

[0032] Figure 1B This is another implementation environment diagram of the teaching evaluation information generation method provided in the embodiment of the present application. Figure 1B , a teacher camera 12, a student camera 11 and a recording and broadcasting host 13 are set in the classroom. Among them, the teacher camera 12 is located above the back wall of the classroom and is used to shoot the teacher area 14. The student camera 11 is located above the podium and is used to shoot the student area 15. The area above the dotted line in the figure including the podium is the teacher area 14, and the area below the dotted line is the student area 15.

[0033] The teacher camera 12 and the student camera 11 can be binocular cameras with close-up and panoramic lenses, and can also be equipped with a pan / tilt system to facilitate intelligent tracking of classroom behavior. The teacher camera 12 and the student camera 11 are both connected to the recording and broadcasting host 13 through a network.

[0034] During the lecture, the recording and broadcasting host 13 pulls the teacher's panoramic stream from the teacher camera 12, and pulls the student's panoramic stream from the student camera 11, and generates a first heat map in real time based on the student's panoramic stream, and generates a second heat map in real time based on the teacher's panoramic stream. By fusing the first heat map and the second heat map, a target heat map reflecting the target teacher's behavior trajectory can be obtained, and teaching evaluation can be performed from an interactive perspective based on the target heat map.

[0035] It should be noted that Figure 1A and Figure 1BIn addition to the teacher camera 12, the student camera 11 and the recording and broadcasting host 13, a loudspeaker, a teacher microphone, a student microphone, a touch-control integrated machine, an interactive TV, an audio box, etc. are also provided. The touch-control integrated machine can be, for example, an interactive tablet, also known as an electronic blackboard, etc. The loudspeaker is used to amplify the teacher's voice, play the sound from the remote end, etc. The teacher's microphone can be, for example, a wireless microphone, supporting various forms such as handheld, collar clip, neck hanging, and ear wear. The student microphone can be, for example, a hanging microphone, a directional microphone, a digital array microphone, etc. An adaptive filter is deployed on the audio box to perform echo cancellation on the audio signals collected by each microphone. The audio box can be integrated on the recording and broadcasting host 13, or it can be independent of the recording and broadcasting host 13, and the embodiments of the present application are not limited.

[0036] Below, based on Figure 1A and Figure 1B The implementation environment shown is used to describe in detail the teaching evaluation information generation method described in the embodiment of this application. For example, please refer to Figure 2 , Figure 2 is a flow chart of a method for generating teaching evaluation information provided by an embodiment of the present application. The execution subject of this embodiment is an electronic device, which may be, for example, Figure 1A The recording and broadcasting host in this embodiment includes:

[0037] 201. Pull the student panoramic stream from the student camera.

[0038] Please refer to Figure 1A The student panoramic stream is a Real-Time Streaming Protocol (RTSP) stream of the student area captured by the student camera.

[0039] 202. Generate a first heat map based on the student panoramic flow, wherein the first heat map is used to indicate the movement trajectory of the target teacher in the student area. The classroom where the target teacher is located includes a teacher area and the student area, and the student area is the field of view of the student camera.

[0040] The student panoramic stream contains a plurality of continuous image frames. The electronic device pulls the student panoramic stream from the student camera, which is equivalent to continuously pulling frames of images from the student camera in real time. For the sake of clarity, the image frame in the student panoramic stream is referred to as the first image frame below.

[0041] Initially, the electronic device creates a blank first heat map, which can be, for example, an image of the same size as the first image frame, or the first heat map has the same size as the projection image of the first image frame after being projected into a bird's-eye view (BEV) coordinate system. Afterwards, each time the first image frame is pulled from the student camera, if there is a target teacher in the first image frame, a scattered point representing the target teacher is rendered in the first heat map according to the position of the target teacher in the first image frame.

[0042] It is understandable that if the target teacher is at the podium throughout the class and does not walk down from the podium to the student area, then the target teacher does not exist in each first image frame and there are no scattered points in the first heat map.

[0043] It can also be understood that if the target teacher is in the student area for a period of time but has not moved, then during this period, the position of the target teacher in each first image frame pulled by the electronic device from the student camera is the same, so that the scattered points representing the target teacher rendered by the electronic device in the first thermal map overlap, that is, the number of scattered points in the first thermal map will not increase. If the target teacher moves frequently, the position of the target teacher in some first image frames is different. Correspondingly, the position of the scattered points representing the target teacher rendered by the electronic device in the first thermal map is also different.

[0044] The embodiment of the present application does not limit the manner in which the electronic device identifies the target teacher from the first image frame.

[0045] For example, facial features of a target teacher are pre-stored on the electronic device. Each time the electronic device pulls the first image frame, it performs multi-target detection on the first image frame to detect the face contained in the first image frame. When there are multiple faces in the first image frame, each face is compared with the pre-stored facial features of the target teacher. If the target teacher's face exists among the multiple faces, a scatter point is rendered in the first heat map.

[0046] For another example, during class, the target teacher comes to the student area. The electronic device determines the object that appears the most times and is standing from the first image frame pulled in the initial 1 minute, and uses the object as the target teacher.

[0047] 203. Generate teaching evaluation information of the target teacher according to the first heat map.

[0048] The electronic device generates teaching evaluation information of the target teacher according to the number and density of scattered points in the first heat map. The teaching evaluation information is used to indicate the interaction between the target teacher and the students. For example, when the number of scattered points in the first heat map is greater than the preset number, the electronic device generates the following teaching evaluation information: the target teacher comes to the student area and stays in the student area for a long time, and interacts more with the students.

[0049] For another example, when the number of scattered points in the first heat map is relatively small, less than or equal to a preset number, the electronic device generates the following teaching evaluation information: although the target teacher came to the student area, he did not stay there for long and had little interaction with the students.

[0050] For another example, in the first heat map, the scattered points are concentrated in a certain area and are relatively dense. The electronic device generates the following teaching evaluation information: the target teacher has more interactions with the students in this area.

[0051] The teaching evaluation information generation method provided in the embodiment of the present application is that the electronic device pulls the student panoramic stream from the student camera, generates a first heat map based on the student panoramic stream, and generates the teaching evaluation information of the target teacher based on the first heat map. Among them, the first heat map is used to indicate the movement trajectory of the target teacher in the student area. The classroom where the target teacher is located includes a teacher area and a student area. The student area is the field of view of the student camera, and the teaching evaluation information is used to indicate the interaction between the target teacher and the student. With this scheme, by generating a real-time first heat map for reflecting the movement trajectory of the target teacher in the student area, the teaching evaluation information of the target teacher is generated based on the first heat map, so as to conduct teaching evaluation from the perspective of interaction, and achieve the purpose of improving the comprehensiveness of the teaching evaluation results.

[0052] Figure 3 is another flow chart of the teaching evaluation information generation method provided by the embodiment of the present application. The execution subject of this embodiment is an electronic device, which can be, for example, Figure 1B The method includes:

[0053] 301. Pull the student panoramic stream from the student camera.

[0054] 302. Pull the teacher's panoramic stream from the teacher's camera.

[0055] In the embodiment of the present application, the electronic device simultaneously pulls a real-time teacher panoramic stream from the teacher camera and a real-time student panoramic stream from the student camera. The teacher panoramic stream is a Real Time Streaming Protocol (RTSP) stream of the teacher area shot by the teacher camera, and the student panoramic stream is an RTSP stream of the student area shot by the student camera.

[0056] The teacher's panoramic stream contains multiple continuous image frames. The electronic device pulls the teacher's panoramic stream from the teacher's camera, which is equivalent to continuously pulling frames of images from the teacher's camera in real time. Similarly, the student's panoramic stream also contains multiple continuous image frames. The electronic device pulls the student's panoramic stream from the student's camera, which is equivalent to continuously pulling frames of images from the student's camera in real time. For the sake of clarity, the image frame in the student's panoramic stream is referred to as the first image frame. The image frame corresponding to the teacher's panoramic stream is referred to as the second image frame.

[0057] It should be noted that there is no strict sequence between step 301 and step 302, but they can be executed in parallel.

[0058] 303. Generate a first heat map based on the student panoramic flow, where the first heat map is used to indicate the movement trajectory of the target teacher in the student area.

[0059] During initialization, the electronic device creates a blank first heat map. After that, each time the electronic device pulls the first image frame, it updates the first heat map. The first heat map is a scatter plot, and each scatter point represents a target teacher. After a class ends or the recording stops, the scatter points on the first heat map represent the target teacher's behavior trajectory in the student area. The more scatter points there are, the more times the target teacher moves in the student area. The denser the scatter points, the smaller the target teacher's range of movement in the student area. A large number of scatter points and sparse scatter points indicate a large range of movement of the target teacher in the student area.

[0060] 304. Generate a second heat map based on the teacher panoramic flow, where the second heat map is used to indicate the movement trajectory of the target teacher in the teacher area.

[0061] Similarly, initially, the electronic device creates a blank second heat map. After that, the second heat map is continuously and dynamically updated. That is, each time the electronic device pulls the second image frame, it updates the second heat map, which is also called a scatter plot, and each point represents the target teacher. After a class ends or the recording stops, the scatter points on the second heat map represent the behavior trajectory of the target teacher in the teacher area. The more scatter points there are, the more times the target teacher walks in the teacher area. The denser the scatter points, the smaller the target teacher's range of movement in the teacher area, and the sparser the scatter points, the larger the target teacher's range of movement in the teacher area.

[0062] 305. The first thermodynamic map and the second thermodynamic map are integrated to obtain a target thermodynamic map.

[0063] After the electronic device generates the first thermal map and the second thermal map in real time, it fuses the first thermal map and the second thermal map to obtain a real-time target thermal map. It is understandable that since the first thermal map and the second thermal map change dynamically, the target thermal map also changes dynamically. The target thermal map at the same time is formed by the fusion of the first thermal map and the second thermal map at the same time.

[0064] 306. Generate teaching evaluation information of the target teacher according to the target heat map.

[0065] The electronic device compares the student area and the teacher area in the target heat map, and generates the teaching evaluation information of the target teacher based on the comparison results. For example, there are no scattered points in the student area, and the electronic device generates the following teaching evaluation information: the target teacher has never walked down from the podium to the student area to interact with students during class, and there is little classroom interaction; for another example, there are a small number of scattered points in the student area and a large number of scattered points in the teacher area, and the electronic device generates the following teaching evaluation information: although the target teacher came to the student area to interact during class, the number of times was relatively small. For another example, there are a large number of scattered points in the student area and a small number of scattered points in the teacher area, and the electronic device generates the following teaching evaluation information: the target teacher came to the student area many times during class and interacted, and the number of interactions was large.

[0066] Figure 4 This is the target heat map in the teaching evaluation information generation method provided in the embodiment of this application. Figure 4 , the target teacher's moving track in the teacher area is mainly concentrated in the podium area, as shown in the dotted rectangle in the figure. The target teacher's moving track in the student area is mainly concentrated in the left aisle. Figure 4 It can be seen that during the class, the target teacher interacted more with the student on the left.

[0067] During the teaching evaluation process, the teaching evaluation information of the target teacher can be generated based on the real-time target heat map, or the teaching evaluation information of the target teacher can be generated based on the final target heat map after the get out of class is over or the recording is stopped. The embodiments of the present application are not limited thereto.

[0068] The teaching evaluation information generation method provided in the embodiment of the present application is that the electronic device pulls the teacher's panoramic stream from the teacher's camera, pulls the student's panoramic stream from the student's camera, generates a first heat map based on the student's panoramic stream, generates a second heat map based on the teacher's panoramic stream, fuses the first heat map and the second heat map to obtain a target heat map, and generates the teaching evaluation information of the target teacher based on the target heat map. With this scheme, by generating a real-time target heat map that reflects the movement trajectory of the target teacher in the student area and the teacher area, comparing the movement trajectory of the target teacher in the student area and the teacher area in the target heat map, and generating the teaching evaluation information of the target teacher, the teaching evaluation is performed from an interactive perspective, achieving the purpose of improving the comprehensiveness of the teaching evaluation results.

[0069] Optionally, in the above embodiment, the student panoramic stream includes a plurality of continuous first image frames, and in the process of the electronic device generating the first heat map according to the student panoramic stream, for each first image frame of the student panoramic stream, a plurality of first candidate frames are determined according to the first image frame, and the objects in the first candidate frames are standing objects or moving objects. Afterwards, when there is a first candidate frame including the target teacher among the plurality of first candidate frames, the first image frame is projected to a bird's-eye view coordinate system to obtain a projected image; and according to the position coordinates of the target teacher in the projected image, a rendering position of a scattered point representing the target teacher is determined in the first heat map, and the scattered point is rendered at the rendering position to generate the first heat map.

[0070] In the embodiment of the present application, the electronic device is a device with built-in artificial intelligence (AI) computing power, and the AI ​​algorithm model is deployed on the electronic device, and the AI ​​model is also called a backbone model, a neural network model, etc. The AI ​​algorithm model is used for target detection and key point detection, etc., and can perform target detection and key point detection on the first image frame and the second image frame.

[0071] For each first image frame of the student's panoramic stream, the electronic device performs target detection and key point detection on the first image frame, thereby determining a human frame from the first image frame, and each human frame at least contains the head of the person. The electronic device performs multi-target tracking and records the key point status of each object. The key points refer to the nose, left eye, right eye, left ear, right ear, left shoulder, right shoulder, left elbow, right elbow, left wrist, right wrist, left hip, etc.

[0072] Afterwards, the electronic device selects a first candidate frame from these human frames according to the key point status of the object in each human frame, and the object in the first candidate frame is a standing object or a moving object. After the electronic device determines multiple first candidate frames from the first image frame, it further determines whether there is a first candidate frame containing a target teacher in these first candidate frames. The first candidate frame selects an area from the first image frame, and the selected area contains a standing object or a moving object. If a standing object or a moving object in a first candidate frame is a target teacher, then the first candidate frame is a candidate frame containing the target teacher.

[0073] When the electronic device determines whether there is a first candidate frame containing a target teacher among multiple first candidate frames, for each first candidate frame, the electronic device extracts facial features, re-identification (re-identification, re-ID) features, etc. to obtain a first feature, and compares the first feature with the target feature of the target teacher. Among them, the re-ID feature is used for target recognition in cross-lens scenes, and the re-ID feature includes character clothing, body shape, hairstyle, etc. If the similarity between the first feature and the target feature is very high, it is considered that the object in the first candidate frame is the target teacher, that is, the first candidate frame is the first candidate frame containing the target teacher. For example, the electronic device calculates the distance between the first vector corresponding to the first feature and the target vector corresponding to the target feature. When the distance is less than the preset distance, it means that the object in the first candidate frame is the target teacher.

[0074] The electronic device can flexibly obtain the target features of the target teacher. In one method, the target features of the target teacher are pre-stored on the electronic device. In another method, after the electronic device determines the target teacher based on the initial image frame in the teacher's panoramic stream, it extracts the target teacher's facial features, re-identification features, etc. to obtain the target features.

[0075] In the embodiment of the present application, the teacher's camera is facing the blackboard on the podium to shoot, that is, the shooting angle is a straight shot. The student camera is located above the podium, which is equivalent to an oblique shot. There are many objects in the student area. In order to make the first thermal map more intuitive and accurate, and the display effect is better, it is necessary to project the first image frame to a bird's-eye view coordinate system to obtain a projected image. Therefore, when there is a first candidate frame containing a target teacher among multiple first candidate frames, the electronic device projects the first image frame to a bird's-eye view coordinate system to obtain a projected image. For example, the electronic device determines a homography projection matrix based on the four vertices of the first image frame and the four vertices of the bird's-eye view coordinate system, referred to as a homography matrix or projection matrix. Afterwards, the electronic device projects the first image frame to a bird's-eye view coordinate system according to the homography matrix to obtain a projected image.

[0076] After the electronic device obtains the projected image, the electronic device determines the position coordinates of the target teacher in the projected image based on the position coordinates of the first candidate box corresponding to the target teacher in the first image frame. Since the first heat map and the projected image have the same size, the pixels in the first image frame correspond to the pixels in the projected image one-to-one. Therefore, the electronic device can determine the rendering position of the scattered points representing the target teacher in the first heat map based on the position coordinates of the target teacher in the projected image, and then render the scattered points at the rendering position to generate the first heat map.

[0077] By adopting this scheme, the projection image is obtained by projecting the first image frame into the bird's-eye view coordinate system, and the first heat map is updated according to the position coordinates of the target teacher in the projection image, so that the first heat map is more intuitive, thereby improving the accuracy and display effect of the first heat map and achieving the purpose of improving the comprehensiveness of the teaching evaluation results.

[0078] The above focuses on how to generate the first heat map based on the student panoramic flow. Next, how to generate the second heat map based on the teacher panoramic flow is described in detail.

[0079] Optionally, in the above embodiment, the teacher panoramic stream includes a plurality of continuous second image frames, and the initial image frame is a preset number of second image frames that are ranked first among the plurality of continuous second image frames. The electronic device determines the target teacher based on the initial image frame. Afterwards, the electronic device performs single target tracking on the remaining second image frames in the teacher panoramic stream except the initial image frame. When the target teacher exists in the second image frame, the position coordinates of the target teacher in the second image frame are determined; based on the position coordinates of the target teacher in the second image frame, the rendering position of the scattered points representing the target teacher is determined in the second heat map, and the scattered points are rendered at the rendering position to generate the second heat map.

[0080] Initially, the electronic device creates a blank second heat map. After the class starts, the electronic device determines the target teacher based on a certain number of initial image frames pulled from the teacher's camera. For example, starting from the first second image frame pulled into the teacher's panoramic stream, the second image frame pulled within 2 seconds, the second image frame pulled within 5 seconds, etc. are used as the initial image frame.

[0081] After determining the target teacher, single target tracking is performed on each second image frame after the initial image frame to determine whether the target teacher exists in the second image frame. When the target teacher exists in the second image frame, the electronic device records the key point status of the target teacher in real time and determines the position coordinates of the target teacher in the second image frame. Afterwards, since the second heat map and the second image frame are the same size, the pixels in the second heat map and the pixels in the second image frame correspond one to one. Therefore, the electronic device can determine the rendering position of the scattered points representing the target teacher in the second heat map according to the position coordinates of the target teacher in the second image frame, and then render the scattered points at the rendering position. When the target teacher does not exist in the second image frame, the electronic device does not update the second heat map.

[0082] It can be understood that if the target teacher is located in the teacher area and does not move within a period of time, the position coordinates of the target teacher in each second image frame during this period are the same, so that the positions of the scattered points representing the target teacher in the second heat map are the same, that is, the scattered points overlap, so that the number of scattered points in the second heat map will not increase.

[0083] It can also be understood that, since the target teacher is either in the teacher area or in the student area at the same time, there will not be a phenomenon in which the scattered points representing the target teacher are rendered in both the first heat map and the second heat map at the same time.

[0084] With this solution, the electronic device performs single target tracking on each second image frame according to the target teacher. Since each second image frame is only checked to see whether it contains the target teacher, and the second heat map is updated when the target teacher is tracked, the purpose of quickly and accurately generating the second heat map is achieved.

[0085] Next, how the electronic device determines the target teacher according to the initial image frame is described in detail.

[0086] In one embodiment, the number of initial image frames is a preset number, and a second candidate frame is determined from each initial image frame to obtain multiple second candidate frames, each of which contains an object. Afterwards, the electronic device determines the object with the most appearances from the multiple second candidate frames, and uses the object with the most appearances as the target teacher.

[0087] In this method, the electronic device pulls the teacher's panoramic stream from the teacher's camera, and the second image frame in the first few seconds is called the initial image frame. The electronic device determines the target teacher according to the business rules and the initial image frame. For example, the business rule is: the object that appears the most times in multiple initial image frames is the target teacher.

[0088] Usually, there is only one teacher in the podium area during class, so there is only one object in each of the multiple initial image frames. The electronic device determines a second candidate frame from each initial image frame, and each second candidate frame contains the object. Therefore, the electronic device uses the object as the target teacher.

[0089] Sometimes, during class, several students are called to the podium to clean the blackboard, collect homework, etc.; or other teachers discuss with the target teacher and leave, so that the electronic device determines multiple second candidate frames from each initial image frame, some of which are teachers, and some of which are students. The electronic device counts the number of times each object appears in the second candidate frame of each initial image frame, and takes the object with the largest number of appearances as the target teacher.

[0090] After the electronic device determines the target teacher, it extracts features from the target teacher to obtain the target features. In the process of generating the first heat map based on the student panoramic flow, the electronic device determines the first candidate frame from the first image frame and extracts the first feature from the object in the first candidate frame. Afterwards, the electronic device determines whether the object in the first candidate frame is the target teacher based on the first feature and the target feature. For example, the electronic device calculates the distance between the first vector corresponding to the first feature and the target vector corresponding to the target feature. When the distance is less than the preset distance, it indicates that the object in the first candidate frame is the target teacher. When the distance is greater than or equal to the preset distance, it indicates that the object in the first candidate frame is not the target teacher.

[0091] After the electronic device determines the target teacher, it extracts the target teacher's facial features and re-ID features, generates and stores the target features based on the facial features and re-ID features. In this way, during subsequent classes, the electronic device directly determines the target teacher based on the target features and the initial image frame without using business rules.

[0092] By adopting this solution, the electronic device extracts a second candidate frame from each initial image frame. The objects in each second candidate frame may be the same or different. The electronic device simulates a normal class scene and determines the target teacher based on the object that appears most frequently, thereby achieving the purpose of quickly and accurately determining the target teacher.

[0093] In another way, in the process of determining the target teacher according to the initial image frame in the teacher panoramic stream, the electronic device determines a second candidate frame from each initial image frame to obtain multiple second candidate frames, each of which contains an object. Afterwards, the electronic device extracts the second feature of the object in the second candidate frame for each second candidate frame in the multiple second candidate frames; and determines the similarity between the second feature and the pre-stored target feature of the target teacher. When the similarity between the second feature and the target feature is greater than the preset similarity, the object in the second candidate frame is determined to be the target teacher.

[0094] The premise of this method is that the target features of the target teacher are pre-stored on the electronic device. When the target teacher needs to be determined, a second candidate frame is extracted from the initial image frame. If there are multiple second candidate frames, each second candidate frame extracts the second feature, and compares the similarity between the target feature and the second feature. When the similarity between the second feature and the target feature is greater than the preset similarity, the object in the second candidate frame is determined to be the target teacher. When the similarity between the second feature and the target feature is less than or equal to the preset similarity, it is determined that the object in the second candidate frame is not the target teacher.

[0095] With this solution, the target features of the target teacher are pre-stored on the electronic device. When teaching evaluation is required, the electronic device determines the target teacher based on the initial image frame and the target features, thereby achieving the purpose of accurately and quickly determining the target teacher.

[0096] Optionally, in the above embodiment, when the field of view of the student camera and the field of view of the teacher camera intersect, the first image frame and the second image frame at the same moment intersect, so that the first heat map and the second heat map have an intersection area. When fusing the first heat map and the second heat map, the electronic device deletes the scattered points in the second heat map located in the intersection area, and fuses the first heat map and the second heat map after deleting the scattered points in the intersection area to obtain the target heat map. In this way, the intersection area only has scattered points on the first heat map, and no scattered points on the second heat map, avoiding the appearance of noise scattered points in the intersection area. Noise scattered points refer to: a scattered point in the first heat map and a scattered point in the second heat map represent the target teacher at the same moment, but due to reasons such as camera calibration accuracy, the position coordinates of the scattered point in the first heat map and the second heat map are different, resulting in two scattered points in the intersection area.

[0097] In addition, the reason why the scattered points in the first heat map are retained for the intersection area and the scattered points in the second heat map are deleted is because the first heat map is a heat map under the bird's-eye view coordinate system, which can more intuitively reflect the movement trajectory of the target teacher.

[0098] With this solution, when the first heat map and the second heat map intersect, for the intersection area, only the scattered points in the first heat map are retained and the scattered points in the second heat map are deleted, ensuring that no noise scattered points appear in the intersection area, thereby achieving the purpose of improving the fusion quality of the first heat map and the second heat map.

[0099] Figure 5 This is a process diagram of the teaching evaluation information generation method provided in the embodiment of the present application. Figure 5 , this embodiment includes:

[0100] 501. Pull the student panoramic stream from the student camera.

[0101] 502. Perform multi-task detection each time after the first image frame is pulled.

[0102] The electronic device is a device with built-in AI computing power, on which an AI algorithm model is deployed, for example, a multi-task detection algorithm model. The electronic device runs the AI ​​algorithm model, takes the first image frame as the input of the AI ​​algorithm model, performs target detection and key point detection on the first image frame, so that the AI ​​algorithm model outputs the human body frame and key point information of the object in the first image frame.

[0103] 503. Perform multi-target tracking on the first image frame.

[0104] The electronic device performs multi-target tracking on the targets detected in step 502 and records the key point status of each human target.

[0105] 504. Determine a first candidate frame containing a standing or moving object from the first image frame.

[0106] The electronic device selects a first candidate frame containing a standing or moving object according to the key point status of each human body frame.

[0107] 505. Extract facial features from the first candidate frame.

[0108] The electronic device performs face frame detection and face feature extraction on the first candidate frame.

[0109] 506. Extract reid features for the first candidate box.

[0110] 507. Determine that the object in the first candidate box is the target teacher.

[0111] The electronic device obtains the first feature of the first candidate frame according to the facial feature and / or the reid feature, and performs distance calculation on the first feature and the target feature, etc., to identify whether the object in the first candidate frame is the target teacher. If the object in the first candidate frame is the target teacher, step 508 is executed; if the object in the first candidate frame is not the target teacher, step 501 is returned. The electronic device obtains the target feature through step 515.

[0112] 508. Project the first image frame to a bird's-eye view coordinate system to obtain a projected image.

[0113] 509. Generate a first heat map according to the projection image and the position coordinates of the target teacher in the projection image.

[0114] In the embodiment of the present application, since the first heat map and the projected image have the same size, the pixels in the first image frame and the pixels in the projected image correspond one to one. Therefore, the electronic device can determine the rendering position of the scattered points representing the target teacher in the first heat map according to the position coordinates of the target teacher in the projected image, and then render the scattered points at the rendering position to generate the first heat map.

[0115] 510. Pull the teacher's panoramic stream from the teacher's camera.

[0116] 511. Perform multi-task detection on the initial image frame.

[0117] The electronic device may be a device with built-in AI computing power, on which an AI algorithm model is deployed, for example, a multi-task detection algorithm model. The electronic device runs the AI ​​algorithm model and uses the initial image frame as the input of the AI ​​algorithm model, so that the AI ​​algorithm model outputs the human body frame and key point information in the initial image frame.

[0118] 512. Determine target teachers based on business rules.

[0119] In step 511, the electronic device performs multi-task detection on each initial image frame to obtain the second candidate frame and key point information. The electronic device determines the target teacher from the multiple second candidate frames according to the business rules. For example, the electronic device determines the object with the most appearances from the multiple second candidate frames, and uses the object with the most appearances as the target teacher.

[0120] 513. Extract facial features of the target teacher.

[0121] 514. Extract re-id features for target teachers.

[0122] 515. Store target features.

[0123] The electronic device uses the facial features and / or the re-ID features as target features and stores them. The next time a course is recorded, the electronic device directly determines the target teacher based on the target features and the initial image frame.

[0124] 516. Perform single target tracking each time after the second image frame is pulled.

[0125] The electronic device performs single-target tracking on the locked target teacher and records the real-time key point status of the target teacher.

[0126] 517. Generate a second heat map according to the position coordinates of the target teacher in the second image frame.

[0127] In the embodiment of the present application, since the blank second heat map and the second image frame are the same size, the pixels in the second heat map and the pixels in the second image frame correspond one to one. Therefore, the electronic device can determine the rendering position of the scattered points representing the target teacher in the second heat map based on the position coordinates of the target teacher in the second image frame, and then render the scattered points at the rendering position.

[0128] 518. Fuse the first thermodynamic map and the second thermodynamic map to obtain a target thermodynamic map.

[0129] 519. Conduct teaching evaluation based on the target heat map.

[0130] It should be noted that there is no strict order for the above steps. For example, step 510 is not necessarily executed after step 509 is executed. Steps 501 to 509 and steps 510 to 517 can be executed in parallel.

[0131] The first heat map shows the movement trajectory of the target teacher in the student area, which to a certain extent reflects the target teacher's activity in the student area. The second heat map shows the movement trajectory of the target teacher in the teacher area, which to a certain extent reflects the target teacher's activity in the teacher area. Therefore, by fusing the two heat maps, it is possible to evaluate the target teacher's teaching from an interactive perspective.

[0132] In addition, in addition to conducting teaching evaluation based on the target heat map, it is also possible to analyze the target teacher's level of activity in the classroom in real time based on the real-time director video, making the generated teaching evaluation information more comprehensive.

[0133] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0134] Figure 6 The schematic diagram of a teaching evaluation information generating device provided in an embodiment of the present application is as follows. The teaching evaluation information generating device 600 comprises: a pulling module 61 , a processing module 62 and a generating module 63 .

[0135] A pulling module 61 is used to pull a student panoramic stream from a student camera;

[0136] A processing module 62 is used to generate a first heat map according to the student panoramic flow, wherein the first heat map is used to indicate a movement trajectory of a target teacher in a student area, wherein the classroom where the target teacher is located includes a teacher area and the student area, and the student area is a field of view of the student camera;

[0137] The generating module 63 is used to generate the teaching evaluation information of the target teacher according to the first heat map.

[0138] In a feasible implementation, the pulling module 61 is also used to pull the teacher's panoramic stream from the teacher's camera;

[0139] The processing module 62 is further used to generate a second heat map according to the teacher panoramic flow, wherein the second heat map is used to indicate the movement trajectory of the target teacher in the teacher area; and to fuse the first heat map and the second heat map to obtain a target heat map;

[0140] The generating module 63 is used to generate the teaching evaluation information of the target teacher according to the target heat map.

[0141] In a feasible implementation, the student panoramic stream includes a plurality of continuous first image frames, and the processing module 62 is used to determine a plurality of first candidate frames according to each first image frame of the student panoramic stream, and the objects in the first candidate frames are standing objects or moving objects; when there is a first candidate frame including the target teacher among the plurality of first candidate frames, the first image frame is projected to a bird's-eye view coordinate system to obtain a projected image; according to the position coordinates of the target teacher in the projected image, a rendering position of a scattered point representing the target teacher is determined in the first heat map, and the scattered point is rendered at the rendering position to generate the first heat map.

[0142] In a feasible implementation, the processing module 62 is specifically used to determine the target teacher based on a preset number of initial image frames in the teacher panoramic stream, the teacher panoramic stream comprising a plurality of consecutive second image frames, the initial image frames being a preset number of second image frames that are ranked first in the plurality of consecutive second image frames, and perform single target tracking on the remaining second image frames in the teacher panoramic stream except the initial image frames; when the target teacher exists in the second image frame, determine the position coordinates of the target teacher in the second image frame; according to the position coordinates of the target teacher in the second image frame, determine the rendering position of the scattered points representing the target teacher in the second heat map, and render the scattered points at the rendering position to generate the second heat map.

[0143] In a feasible implementation, the processing module 62 is specifically used to determine the target teacher based on the initial image frame in the teacher panoramic stream, and to determine a second candidate frame from each initial image frame to obtain multiple second candidate frames, each second candidate frame containing an object; determine the object with the most appearances from the multiple second candidate frames, and use the object with the most appearances as the target teacher.

[0144] In a feasible implementation, the number of the initial image frames is a preset number, and the processing module 62 is specifically used to determine the target teacher based on the initial image frames in the teacher panoramic stream, and to determine a second candidate frame from each initial image frame to obtain multiple second candidate frames, each second candidate frame containing an object; for each second candidate frame in the multiple second candidate frames, extracting a second feature of the object in the second candidate frame; determining the similarity between the second feature and a pre-stored target feature of the target teacher; when the similarity between the second feature and the target feature is greater than the preset similarity, determining that the object in the second candidate frame is the target teacher.

[0145] In a feasible implementation, the processing module 62 is specifically used to fuse the first heat map and the second heat map to obtain a target heat map. When the first heat map and the second heat map have an intersection area, the scattered points in the second heat map located in the intersection area are deleted, and the second heat map after deleting the scattered points in the intersection area and the first heat map are fused to obtain the target heat map.

[0146] The teaching evaluation information generating device provided in the embodiment of the present application can execute the actions of the electronic device in the above embodiment. Its implementation principle and technical effect are similar and will not be repeated here.

[0147] Figure 7 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application, the electronic device is, for example, the above-mentioned robot, etc. Figure 7 The electronic device 700 described in the embodiment of the present application includes: at least one processor 71, at least one communication bus 72, a user interface 73, at least one network interface 74 and a memory 75.

[0148] The communication bus 72 is used to realize the connection and communication between these components.

[0149] The user interface 73 may include a display screen (Display) and a camera (Camera), and the optional user interface 73 may also include a standard wired interface and a wireless interface.

[0150] The network interface 74 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0151] Among them, the processor 71 may include one or more processing cores. The processor 71 uses various interfaces and lines to connect various parts of the entire electronic device 700, and executes various functions and processes data of the electronic device 700 by running or executing instructions, programs, code sets or instruction sets stored in the memory 75, and calling data stored in the memory 75. Optionally, the processor 71 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 71 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for the rendering and drawing of the panoramic ball required to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 71, and can be implemented separately through a chip.

[0152] Among them, the memory 75 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 75 includes a non-transitory computer-readable storage medium. The memory 75 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 75 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 75 may also be optionally at least one storage device located away from the aforementioned processor 71. As Figure 7 As shown, the memory 75 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an operating application of the electronic device.

[0153] An embodiment of the present application also provides a computer-readable storage medium, in which computer instructions are stored. When the computer instructions are executed by a processor, they are used to implement the teaching evaluation information generation method as described above.

[0154] An embodiment of the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the teaching evaluation information generation method as described above is implemented.

[0155] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0156] The present application is described with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, and the combination of the process and / or box in the flowchart and / or block diagram can be realized by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for realizing the function specified in one process or multiple processes in the flowchart and / or one box or multiple boxes in the block diagram.

[0157] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0158] These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0159] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0160] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0161] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0162] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0163] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for generating teaching evaluation information, characterized in that: Applied to electronic equipment, the method comprises: Pull the student panoramic stream from the student camera; Generate a first heat map according to the student panoramic flow, wherein the first heat map is used to indicate a movement trajectory of a target teacher in a student area, wherein the classroom where the target teacher is located includes a teacher area and the student area, and the student area is a field of view of the student camera; The teaching evaluation information of the target teacher is generated according to the first heat map.

2. The method according to claim 1, characterized in that Generating the teaching evaluation information of the target teacher according to the first heat map includes: Pull the teacher's panoramic stream from the teacher's camera; generating a second heat map according to the teacher panoramic flow, wherein the second heat map is used to indicate a moving trajectory of a target teacher in the teacher area; Fusion of the first thermal map and the second thermal map to obtain a target thermal map; The teaching evaluation information of the target teacher is generated according to the target heat map.

3. The method according to claim 1, characterized in that The student panoramic stream includes a plurality of continuous first image frames, and generating a first heat map according to the student panoramic stream includes: For each first image frame of the student panoramic stream, a plurality of first candidate frames are determined according to the first image frame, and the objects in the first candidate frames are standing objects or moving objects; When a first candidate frame including the target teacher exists in the plurality of first candidate frames, projecting the first image frame to a bird's-eye view coordinate system to obtain a projected image; According to the position coordinates of the target teacher in the projection image, the rendering positions of the scattered points representing the target teacher are determined in the first heat map, and the scattered points are rendered at the rendering positions to generate the first heat map.

4. The method according to claim 2, characterized in that: The generating a second heat map according to the teacher panoramic flow includes: Determine the target teacher according to a preset number of initial image frames in the teacher panoramic stream, wherein the teacher panoramic stream includes a plurality of continuous second image frames, and the initial image frames are a preset number of second image frames that are ranked first among the plurality of continuous second image frames; Performing single target tracking on the remaining second image frames in the teacher panoramic stream except the initial image frame; When the target teacher exists in the second image frame, determining the position coordinates of the target teacher in the second image frame; According to the position coordinates of the target teacher in the second image frame, the rendering positions of the scattered points representing the target teacher are determined in the second heat map, and the scattered points are rendered at the rendering positions to generate the second heat map.

5. The method according to claim 4, characterized in that The step of determining the target teacher according to the initial image frame in the teacher panoramic stream comprises: Determine a second candidate frame from each initial image frame to obtain a plurality of second candidate frames, each second candidate frame containing an object; The object that appears most frequently is determined from the multiple second candidate frames, and the object that appears most frequently is used as the target teacher.

6. The method according to claim 4, characterized in that The number of the initial image frames is a preset number, and determining the target teacher according to the initial image frames in the teacher panoramic stream includes: Determine a second candidate frame from each initial image frame to obtain a plurality of second candidate frames, each second candidate frame containing an object; For each second candidate box in the plurality of second candidate boxes, extract a second feature of the object in the second candidate box; Determining a similarity between the second feature and a pre-stored target feature of the target teacher; When the similarity between the second feature and the target feature is greater than a preset similarity, the object in the second candidate frame is determined to be the target teacher.

7. The method according to claim 2 or 4 or 5 or 6, characterized in that: The fusing the first heat map and the second heat map to obtain a target heat map includes: When the first heat map and the second heat map have an intersection area, deleting scattered points in the second heat map located in the intersection area; The second thermodynamic map after deleting the scattered points in the intersection area and the first thermodynamic map are fused to obtain the target thermodynamic map.

8. A teaching evaluation information generating device, characterized in that: include: A pulling module is used to pull the student panoramic stream from the student camera; A processing module, configured to generate a first heat map according to the student panoramic flow, wherein the first heat map is used to indicate a movement trajectory of a target teacher in a student area, wherein a classroom where the target teacher is located includes a teacher area and the student area, and the student area is a field of view of the student camera; A generation module is used to generate teaching evaluation information of the target teacher based on the first heat map.

9. An electronic device comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the electronic device implements the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images