CN113392745A - Abnormal action correction method, abnormal action correction device, electronic equipment and computer storage medium - Google Patents
Abnormal action correction method, abnormal action correction device, electronic equipment and computer storage medium Download PDFInfo
- Publication number
- CN113392745A CN113392745A CN202110626797.6A CN202110626797A CN113392745A CN 113392745 A CN113392745 A CN 113392745A CN 202110626797 A CN202110626797 A CN 202110626797A CN 113392745 A CN113392745 A CN 113392745A
- Authority
- CN
- China
- Prior art keywords
- dimension
- value
- abnormal
- motion
- motion attitude
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 112
- 238000012937 correction Methods 0.000 title claims abstract description 67
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000003860 storage Methods 0.000 title claims abstract description 13
- 230000033001 locomotion Effects 0.000 claims abstract description 205
- 239000013598 vector Substances 0.000 claims abstract description 131
- 238000001514 detection method Methods 0.000 claims abstract description 27
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 19
- 230000009471 action Effects 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 9
- 230000011218 segmentation Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000005856 abnormality Effects 0.000 claims description 4
- 210000000988 bone and bone Anatomy 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 13
- 210000001624 hip Anatomy 0.000 description 7
- 238000004891 communication Methods 0.000 description 6
- 210000004394 hip joint Anatomy 0.000 description 6
- 210000000629 knee joint Anatomy 0.000 description 6
- 210000004197 pelvis Anatomy 0.000 description 6
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 230000036544 posture Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000005452 bending Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 210000002310 elbow joint Anatomy 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 210000000323 shoulder joint Anatomy 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 210000003857 wrist joint Anatomy 0.000 description 2
- 206010000117 Abnormal behaviour Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
Abstract
The embodiment of the application provides an abnormal action correcting method, an abnormal action correcting device, electronic equipment and a computer storage medium. The method comprises the following steps: the method comprises the steps of segmenting a detected video to obtain a motion attitude vector corresponding to each frame of image of the detected video, wherein the motion attitude vector comprises a plurality of dimensional values, and the plurality of dimensional values are used for representing included angles formed by different joint points; detecting whether the dimension value of the motion attitude vector is abnormal or not through an abnormal value detection algorithm; if the motion attitude vector is abnormal, correction data are obtained according to the dimension value of the motion attitude vector; wherein the correction data is used for characterizing adjustment data of the abnormal dimension value. By adopting the abnormal action correcting method provided by the application, the manual experience of professional teachers is not needed, the use is simple and convenient, and the method is suitable for wide popularization and application.
Description
Technical Field
The present application relates to the field of video processing technologies, and in particular, to a method and an apparatus for correcting an abnormal motion, an electronic device, and a computer storage medium.
Background
Scenes such as sports education, dance education and body building under online mainly depend on the manual experience of professional teachers, and the root of the wrong action is searched in a mode of observing the movement posture by naked eyes. For example, in the case of performing a push-up exercise, the waist needs to be kept horizontal, and if the push-up exercise is performed while bending down, an abnormal motion occurs due to the fact that the waist is not kept horizontal.
Problems existing in the prior art:
the root cause analysis of abnormal actions is carried out manually, which not only consumes human resources, but also cannot be widely applied. And professional teachers are limited in number, and can not analyze the root cause of abnormal actions in time aiming at massive movement actions, so that a scheme for correcting the abnormal actions can not be provided in time.
Disclosure of Invention
The embodiment of the application provides an abnormal action correcting method, an abnormal action correcting device, electronic equipment and a computer storage medium, and aims to solve the problems that in the prior art, human resources are consumed, the abnormal action correcting method cannot be widely applied, the root cause of the abnormal action cannot be analyzed in time, and a scheme for correcting the abnormal action cannot be provided in time.
According to a first aspect of embodiments of the present application, there is provided an abnormal motion correction method, including:
dividing a detected video to obtain a motion attitude vector corresponding to each frame of image of the detected video; the motion attitude vector is used for representing a motion attitude, and comprises a plurality of dimension values which are used for representing included angles formed by different joint points;
detecting whether the dimension value of the motion attitude vector is abnormal or not through an abnormal value detection algorithm;
if the motion attitude vector is abnormal, correction data are obtained according to the dimension value of the motion attitude vector; wherein the correction data is used for adjustment data characterizing the dimension value of the anomaly.
According to a second aspect of embodiments of the present application, there is provided an abnormal motion correction apparatus, including:
the segmentation module is used for carrying out segmentation processing on a detected video to obtain a motion attitude vector corresponding to each frame of image of the detected video; the motion attitude vector is used for representing a motion attitude, and comprises a plurality of dimension values which are used for representing included angles formed by different joint points;
the abnormity judgment module is used for detecting whether the dimension value of the motion attitude vector is abnormal or not through an abnormal value detection algorithm;
the correction module is used for obtaining correction data according to the dimension value of the motion attitude vector if the motion attitude vector is abnormal; wherein the correction data is used for adjustment data characterizing the dimension value of the anomaly.
According to a third aspect of embodiments of the present application, there is provided an electronic device comprising one or more processors, and memory for storing one or more programs; the one or more programs, when executed by the one or more processors, implement the steps of the abnormal action correction method as described above.
According to a fourth aspect of embodiments of the present application, there is provided a computer storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the abnormal behavior correction method as described above.
By adopting the abnormal action correcting method, the abnormal action correcting device, the electronic equipment and the computer storage medium provided by the embodiment of the application, the motion attitude vector corresponding to each frame of image of the detected video is obtained by segmenting the detected video, wherein the motion attitude vector comprises a plurality of dimensional values, and the plurality of dimensional values are used for representing included angles formed by different joint points; detecting whether the dimension value of the motion attitude vector is abnormal or not through an abnormal value detection algorithm; if the motion attitude vector is abnormal, correction data are obtained according to the dimension value of the motion attitude vector; wherein the correction data is used for characterizing adjustment data of the abnormal dimension value. Therefore, abnormal actions in the detected video can be rapidly detected through an abnormal value detection algorithm, and correction data can be obtained according to the dimension value of the motion attitude vector. The abnormal action correction method is implemented without depending on the manual experience of professional teachers, is simple and convenient to use, and is suitable for wide popularization and application.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
fig. 1 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure;
fig. 2 is a schematic flowchart of an abnormal operation correcting method according to an embodiment of the present application;
fig. 3 is a schematic flowchart of another abnormal motion correction method according to an embodiment of the present application;
fig. 4 is a schematic flowchart of another abnormal motion correction method according to an embodiment of the present application;
fig. 5 is a schematic flowchart of another abnormal motion correction method according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an abnormal motion correction apparatus according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of another abnormal motion correction apparatus according to an embodiment of the present application.
Detailed Description
In the process of realizing the application, the inventor finds that scenes such as sports education, dance education, fitness and the like under online are mainly based on the manual experience of professional teachers, and the root of the wrong action is searched in a mode of observing the motion posture by naked eyes. For example, in the case of performing a push-up exercise, the waist needs to be kept horizontal, and if the push-up exercise is performed while bending down, an abnormal motion occurs due to the fact that the waist is not kept horizontal. The root cause analysis of abnormal actions is carried out manually, which not only consumes human resources, but also cannot be widely applied. And professional teachers are limited in number, and can not analyze the root cause of abnormal actions in time aiming at massive movement actions, so that a scheme for correcting the abnormal actions can not be provided in time.
In view of the foregoing problems, embodiments of the present application provide a method, an apparatus, an electronic device, and a computer storage medium for correcting an abnormal motion, which can quickly detect an abnormal motion in a detected video through an abnormal value detection algorithm, and obtain correction data according to a dimension value of a motion attitude vector. The abnormal action correction method is implemented without depending on the manual experience of professional teachers, is simple and convenient to use, and is suitable for wide popularization and application.
The scheme in the embodiment of the application can be implemented by adopting various computer languages, such as object-oriented programming language Java and transliterated scripting language JavaScript.
In order to make the technical solutions and advantages of the embodiments of the present application more apparent, the following further detailed description of the exemplary embodiments of the present application with reference to the accompanying drawings makes it clear that the described embodiments are only a part of the embodiments of the present application, and are not exhaustive of all embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
As shown in fig. 1, a schematic structural diagram of an electronic device 100 provided in an embodiment of the present application is shown, where the electronic device 100 includes a memory 101, a processor 102, and a communication interface 103. The memory 101, processor 102 and communication interface 103 are electrically connected to each other, directly or indirectly, to enable the transfer or interaction of data. For example, the components may be electrically connected to each other via one or more communication buses or signal lines. The memory 101 may be used to store software programs and modules, such as program instructions/modules corresponding to the abnormal operation correcting method provided in the embodiment of the present application, and the processor 102 executes the software programs and modules stored in the memory 101, thereby executing various functional applications and data processing. The communication interface 103 may be used for communication of signaling or data with the node apparatus 300 and the client 200. The electronic device 100 may have a plurality of communication interfaces 103 in this application.
The Memory 101 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like.
The processor 102 may be an integrated circuit chip having signal processing capabilities. The processor may be a general-purpose processor including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc.
Next, on the basis of the electronic device 100 shown in fig. 1, an embodiment of the present application provides an abnormal motion correction method, please refer to fig. 2, and fig. 2 is an abnormal motion correction method provided in an embodiment of the present application, where the abnormal motion correction method may include the following steps:
s201, the detection video is segmented, and a motion attitude vector corresponding to each frame of image of the detection video is obtained.
The motion attitude vector is used for representing a motion attitude, the motion attitude vector comprises a plurality of dimension values, and the dimension values are used for representing included angles formed by different joint points.
It should be understood that the detected video may be video data of abnormal motion and/or standard motion. Aiming at the video data of the standard action, the abnormal action correction method can determine that the video data of the standard action has no abnormality; aiming at the video data of the abnormal action, the root cause of the abnormal action can be accurately positioned by the abnormal action correcting method, and the correction data can be obtained.
The movements in the detected video may be sports movements, dance movements, fitness movements, and the like. For example, a high jump action, a ballet action, a flat-bed action, a push-up action, and the like are possible.
S202, detecting whether the dimension value of the motion attitude vector is abnormal or not through an abnormal value detection algorithm.
It should be understood that the outlier detection algorithm can be understood as: the dimension value of the motion attitude vector needs to obey normal distribution, and if the dimension value of the motion attitude vector does not obey the normal distribution, the motion attitude vector is abnormal; and if the dimension value of the motion attitude vector is subjected to normal distribution, the motion attitude vector is normal.
And S203, if the motion attitude vector is abnormal, obtaining correction data according to the dimension value of the motion attitude vector.
Wherein the correction data is used for characterizing adjustment data of the abnormal dimension value. It can be understood that the correction data is an adjustment angle of an included angle formed by target joint points, and the target joint points are corresponding to abnormal dimension values.
The operation in the detection video is taken as a push-up, and if the push-up is performed while bending down, the push-up operation in the detection video is an abnormal operation. The abnormal dimension value in the motion posture vector is an included angle formed by joint points related to the waist, the target joint point is the joint point related to the waist, and the correction data is an adjustment angle of the included angle formed by the joint points related to the waist.
In order to facilitate understanding how to obtain a motion posture vector corresponding to each frame of image of a detected video, please refer to fig. 3, which is another abnormal motion correction method provided in this embodiment of the present application, the above-mentioned S201 includes the following sub-steps:
s201a, detecting the joint points in each frame of video image according to the bone key point detection algorithm.
It should be understood that the three-dimensional coordinates of the joint points of each frame of image in the detection video can be detected by the bone key point detection algorithm. The joint points may include pelvic joints, left shoulder joints, right shoulder joints, left elbow joints, right elbow joints, left wrist joints, right wrist joints, left hip joints, right hip joints, left knee joints, right knee joints, and the like.
And S201b, calculating to obtain a motion attitude vector according to the joint point.
It should be understood that the motion pose vector may be described by the angle formed between the different joint points. One motion attitude vector may be a set of angles formed between all the joint points in one frame image, or may be a set of angles formed between specified joint points in one frame image. The motion postures in different frame images are different, and the included angles formed between the joint points are different in size. But the joint points in each frame of the image are the same.
For example, 3 angles can be formed among all joint points in each frame of image in the detection video, which are respectively a first angle formed among the left knee joint, the left hip joint and the pelvis, a second angle formed among the right knee joint, the right hip joint and the pelvis, and a third angle formed among the vertex, the neck and the pelvis. The motion attitude vector generated corresponding to each frame of image is a set of the first angle, the second angle and the third angle.
In order to facilitate understanding how to detect whether the dimension value of the motion pose vector is abnormal through the abnormal value detection algorithm, please refer to fig. 4, which is a flowchart illustrating another abnormal motion correction method provided in an embodiment of the present application, where the above step S202 includes the following sub-steps:
s202a, matching the dimension value of the motion attitude vector with the threshold range.
And each dimension value is respectively provided with a corresponding threshold range, and the threshold range is used for representing a normal value range of the corresponding dimension value. It can be understood that if the motion attitude vector includes three dimensions, the three dimensions are respectively a first dimension value, a second dimension value and a third dimension value; the threshold range correspondingly comprises three threshold ranges, namely a first threshold range, a second threshold range and a third threshold range; and the first dimension value corresponds to a first threshold range, the second dimension value corresponds to a second threshold range, and the third dimension value corresponds to a third threshold range.
It will be appreciated that the correspondence of the dimension value to the threshold range may be determined from the joint points forming the dimension value. The threshold range is determined according to the target dimension value of the sample motion attitude vector corresponding to each frame image of the standard motion videos.
S202b, if the dimension value of the motion orientation vector is not within the threshold range, it is determined that the dimension value of the motion orientation vector is abnormal.
It should be understood that if the dimension value of the motion attitude vector is within the threshold range, the dimension value of the motion attitude vector is determined to be normal.
And if the dimension value of the motion attitude vector is abnormal, obtaining correction data according to the dimension value of the motion attitude vector and the threshold range. It can be understood that if the dimension value of the motion attitude vector is greater than the upper threshold of the threshold range, the upper threshold is the maximum value in the threshold range; and subtracting the dimensional value of the motion attitude vector from the upper limit threshold value to obtain the correction data. If the dimension value of the motion attitude vector is smaller than the lower threshold of the threshold range, the lower threshold is the minimum value in the threshold range; and subtracting the dimension value of the motion attitude vector from the lower threshold to obtain the correction data.
For example, if the threshold range takes a value of 60 ° to 90 °, the motion attitude vector has a dimension value of 100 °, and the dimension value of 100 ° of the motion attitude vector is subtracted from the upper threshold of 90 ° of the threshold range to obtain 10 ° of correction data. Namely, the included angle of the joint point corresponding to the dimension value of the motion attitude vector is adjusted to be smaller by 10 degrees, so that the abnormal motion in the motion can be corrected in time, the motion efficiency is improved, and the motion damage caused by the abnormal motion is reduced.
To understand how to determine the threshold range, please refer to fig. 5, which is a schematic flowchart of another abnormal motion correction method provided in the embodiment of the present application, and based on the abnormal motion correction method shown in fig. 2, the abnormal motion correction method further includes the following steps:
s301, sample motion attitude vectors corresponding to each frame of image of a plurality of standard motion videos are obtained.
Wherein each sample motion pose vector comprises a plurality of target dimension values; the frame number of each standard action video is the same; the number of the dimension values of each sample motion attitude vector is the same, and the corresponding joint points of each dimension value in different sample motion attitude vectors are respectively and correspondingly the same.
It should be understood that the motions in the plurality of standard motion videos should be the same, for example, the motions in each standard motion video may be push-up standard motions, flat-panel standard motions or ballet standard motions, and the plurality of standard motion videos may be generated by one professional performing a plurality of sets of standard motions repeatedly, or may be generated by a plurality of professionals performing a plurality of sets of standard motions repeatedly.
The implementation principle of obtaining the sample motion attitude vector corresponding to each frame of image in the standard motion video from the standard motion video is the same as the implementation principle of obtaining the motion attitude vector corresponding to each frame of image in the detection video from the detection video. Reference is made to the corresponding contents of S201, S21a and S201b, and the description is not repeated here.
In the embodiment of the application, the number of frames of each standard motion video slice is the same. For example, the duration of each standard motion video is 30s, and each standard motion video may be segmented into 60 frames at a time interval of 0.5s, and a corresponding image of each frame is obtained.
In the embodiment of the present application, the number of target dimension values of each sample motion pose vector should be the same, and the joint points corresponding to the target dimension values of different sample motion pose vectors should also be the same. For example, the number of target dimension values of each sample motion attitude vector is 3, and the target dimension values are respectively a first target dimension value, a second target dimension value and a third target dimension value; the joint points corresponding to the first target dimension value of each sample motion attitude vector are a left knee joint, a left hip joint and a pelvis, the joint points corresponding to the second target dimension value of each sample motion attitude vector are a right knee joint, a right hip joint and a pelvis, and the joint points corresponding to the third target dimension value of each sample motion attitude vector are a vertex, a neck and a pelvis.
S302, calculating the mean value and the standard deviation of the target dimension values corresponding to the same joint points to obtain the mean value and the standard deviation of the dimension.
It should be understood that if there are m standard motion videos, each standard motion video includes n frames of images, and the motion attitude vector corresponding to each frame of image includes L target dimension values; the target dimension values corresponding to the same joint point are subjected to mean value and standard deviation calculation, that is, the mean value and standard deviation calculation is performed on the ith target dimension value in the motion attitude vector of the t-th frame image in each standard motion video, so that the dimension mean value and the dimension standard deviation of the ith dimension in the motion attitude vector of the t-th frame image are obtained. Wherein, the value range of t is 1-m, and the value range of i is 1-L; namely, the t frame image is any one frame in the standard motion video, and the ith target dimension value is any one target dimension value in the motion attitude vector.
The dimension mean value of the ith dimension in the motion attitude vector of the t frame image can be calculated by adopting the following formula:
wherein the content of the first and second substances,
the dimension mean value of the ith dimension in the motion attitude vector of the image of the t-th frame, m is the number of standard motion videos,
and the target dimension value of the ith dimension in the motion attitude vector of the t frame image in the mth standard motion video.
The dimension standard deviation of the ith dimension in the motion attitude vector of the t frame image can be calculated by adopting the following formula:
wherein the content of the first and second substances,
dimension standard deviation of ith dimension in motion attitude vector of t frame image, m is number of standard motion video,
the target dimension value of the ith dimension in the motion attitude vector of the t frame image in the mth standard motion video,
for the motion pose of the t frame imageThe dimension mean of the ith dimension in the vector.
And S303, generating a threshold range according to the dimension mean value and the dimension standard deviation.
It should be understood that subtraction may be performed on the preset multiple of the dimension mean and the dimension standard deviation to obtain a lower threshold; performing addition operation on the preset multiples of the dimension mean value and the dimension standard deviation to obtain an upper limit threshold; a threshold range is generated based on the lower threshold and the upper threshold.
The preset multiple may be set to 3 times, and may be set according to an actual situation, which is not limited herein. The threshold range is an interval value between an upper threshold and a lower threshold; in addition, as can be seen from the above, the threshold ranges corresponding to the dimension values of different dimensions in the motion pose vectors of different frame images may be the same or different.
In order to implement the abnormal operation correcting method corresponding to the above-mentioned S201 to S203, S301 to S303 and possible sub-steps thereof, an embodiment of the present application provides an abnormal operation correcting apparatus, please refer to fig. 6, fig. 6 is a block diagram of an abnormal operation correcting apparatus 400 provided in an embodiment of the present application, where the abnormal operation correcting apparatus 400 includes: a segmentation module 401, an abnormality determination module 402 and a correction module 403.
The segmentation module 401 is configured to segment the detected video to obtain a motion attitude vector corresponding to each frame of image of the detected video; the motion attitude vector is used for representing a motion attitude, the motion attitude vector comprises a plurality of dimension values, and the dimension values are used for representing included angles formed by different joint points.
The anomaly determination module 402 is configured to detect whether the dimension value of the motion attitude vector is abnormal through an anomaly value detection algorithm.
In an optional implementation manner, the anomaly determination module 402 is further configured to match a dimension value of the motion attitude vector with a threshold range; each dimension value is respectively provided with a corresponding threshold range, and the threshold range is used for representing a normal value range of the corresponding dimension value; the anomaly determination module 402 is further configured to determine that the dimension value of the motion attitude vector is abnormal if the dimension value of the motion attitude vector is not within the threshold range.
The correction module 403 is configured to, if the motion gesture vector is abnormal, obtain correction data according to a dimension value of the motion gesture vector; wherein the correction data is used for characterizing adjustment data of the abnormal dimension value.
In an optional implementation manner, the correction module is further configured to obtain correction data according to the dimension value of the motion attitude vector and a threshold range.
In an alternative embodiment, as shown in fig. 7, the abnormal motion correction apparatus 400 further includes: an acquisition module 404, a calculation module 405, and a threshold generation module 406.
The obtaining module 404 is configured to obtain sample motion attitude vectors corresponding to each frame of image of a plurality of standard motion videos; wherein each sample motion pose vector comprises a plurality of target dimension values; the frame number of each standard action video is the same; the number of the dimension values of each sample motion attitude vector is the same, and the corresponding joint points of each dimension value in different sample motion attitude vectors are respectively and correspondingly the same.
The calculation module 405 is configured to calculate a mean value and a standard deviation of the target dimension values corresponding to the same joint point, so as to obtain a dimension mean value and a dimension standard deviation.
The threshold generation module 406 is configured to generate a threshold range according to the dimension mean and the dimension standard deviation.
It should be understood that the segmentation module 401, the abnormality determination module 402, the correction module 403, the acquisition module 404, the calculation module 405, and the threshold generation module 406 may cooperatively implement the above-described S201 to S203, S301 to S303, and possible sub-steps thereof.
In summary, the present application provides a method, an apparatus, an electronic device, and a computer storage medium for correcting an abnormal motion, in which a motion attitude vector corresponding to each frame of image of a detected video is obtained by segmenting the detected video, the motion attitude vector includes a plurality of dimension values, and the plurality of dimension values are used for representing included angles formed by different joint points; detecting whether the dimension value of the motion attitude vector is abnormal or not through an abnormal value detection algorithm; if the motion attitude vector is abnormal, correction data are obtained according to the dimension value of the motion attitude vector; wherein the correction data is used for characterizing adjustment data of the abnormal dimension value. Therefore, abnormal actions in the detected video can be rapidly detected through an abnormal value detection algorithm, and correction data can be obtained according to the dimension value of the motion attitude vector. The abnormal action correction method is implemented without depending on the manual experience of professional teachers, is simple and convenient to use, and is suitable for wide popularization and application.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While the preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.
Claims (10)
1. An abnormal motion correction method, comprising:
dividing a detected video to obtain a motion attitude vector corresponding to each frame of image of the detected video; the motion attitude vector is used for representing a motion attitude, and comprises a plurality of dimension values which are used for representing included angles formed by different joint points;
detecting whether the dimension value of the motion attitude vector is abnormal or not through an abnormal value detection algorithm;
if the motion attitude vector is abnormal, correction data are obtained according to the dimension value of the motion attitude vector; wherein the correction data is used for adjustment data characterizing the dimension value of the anomaly.
2. The method of claim 1, wherein the step of detecting whether the dimension value of the motion pose vector is abnormal by an abnormal value detection algorithm comprises:
matching the dimension value of the motion attitude vector with a threshold range; each dimension value is respectively provided with a corresponding threshold range, and the threshold range is used for representing a normal value range of the corresponding dimension value;
if the dimension value of the motion attitude vector is not within the threshold range, judging that the dimension value of the motion attitude vector is abnormal;
the step of obtaining correction data according to the dimension value of the motion attitude vector comprises:
and obtaining the correction data according to the dimension value of the motion attitude vector and the threshold range.
3. The method of claim 2, wherein before the step of detecting whether the dimension value of the motion pose vector is abnormal by an abnormal value detection algorithm, the method further comprises:
obtaining sample motion attitude vectors corresponding to each frame of image of a plurality of standard motion videos; wherein each of the sample motion pose vectors comprises a plurality of target dimension values; the frame number of each standard action video is the same; the number of the dimension values of each sample motion attitude vector is the same, and the corresponding joint points of each dimension value in different sample motion attitude vectors are respectively correspondingly the same;
calculating the mean value and the standard deviation of target dimension values corresponding to the same joint points to obtain a dimension mean value and a dimension standard deviation;
generating the threshold range from the dimension mean and the dimension standard deviation.
4. The method of claim 3, wherein the step of generating the threshold range from the dimensional mean and the dimensional standard deviation comprises:
carrying out subtraction operation on the dimension mean value and a preset multiple of the dimension standard deviation to obtain a lower limit threshold;
performing addition operation on the dimension mean value and a preset multiple of the dimension standard deviation to obtain an upper limit threshold value;
and generating the threshold range according to the lower threshold and the upper threshold.
5. The method according to any one of claims 1 to 4, wherein the step of performing segmentation processing on the detected video to obtain the motion attitude vector corresponding to each frame of image of the detected video comprises:
detecting the joint points in each frame of image of the detection video according to a bone key point detection algorithm;
and calculating to obtain the motion attitude vector according to the joint point.
6. An abnormal motion correction apparatus, comprising:
the segmentation module is used for carrying out segmentation processing on a detected video to obtain a motion attitude vector corresponding to each frame of image of the detected video; the motion attitude vector is used for representing a motion attitude, and comprises a plurality of dimension values which are used for representing included angles formed by different joint points;
the abnormity judgment module is used for detecting whether the dimension value of the motion attitude vector is abnormal or not through an abnormal value detection algorithm;
the correction module is used for obtaining correction data according to the dimension value of the motion attitude vector if the motion attitude vector is abnormal; wherein the correction data is used for adjustment data characterizing the dimension value of the anomaly.
7. The apparatus of claim 6, wherein the anomaly determination module is further configured to match a dimension value of the motion gesture vector with a threshold range; each dimension value is respectively provided with a corresponding threshold range, and the threshold range is used for representing a normal value range of the corresponding dimension value;
the abnormality judgment module is further configured to judge that the dimension value of the motion attitude vector is abnormal if the dimension value of the motion attitude vector is not within the threshold range;
and the correction module is further used for obtaining the correction data according to the dimension value of the motion attitude vector and the threshold range.
8. The apparatus of claim 7, further comprising
The acquisition module is used for acquiring sample motion attitude vectors corresponding to each frame of image of a plurality of standard motion videos; wherein each of the sample motion pose vectors comprises a plurality of target dimension values; the frame number of each standard action video is the same; the number of the dimension values of each sample motion attitude vector is the same, and the corresponding joint points of each dimension value in different sample motion attitude vectors are respectively correspondingly the same;
the calculation module is used for calculating the mean value and the standard deviation of the target dimension values corresponding to the same joint point to obtain a dimension mean value and a dimension standard deviation;
and the threshold value generation module is used for generating the threshold value range according to the dimension mean value and the dimension standard deviation.
9. An electronic device comprising one or more processors, and memory for storing one or more programs; the one or more programs, when executed by the one or more processors, implement the method of any of claims 1-5.
10. A computer storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110626797.6A CN113392745A (en) | 2021-06-04 | 2021-06-04 | Abnormal action correction method, abnormal action correction device, electronic equipment and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110626797.6A CN113392745A (en) | 2021-06-04 | 2021-06-04 | Abnormal action correction method, abnormal action correction device, electronic equipment and computer storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113392745A true CN113392745A (en) | 2021-09-14 |
Family
ID=77618301
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110626797.6A Pending CN113392745A (en) | 2021-06-04 | 2021-06-04 | Abnormal action correction method, abnormal action correction device, electronic equipment and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113392745A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110163112A (en) * | 2019-04-25 | 2019-08-23 | 沈阳航空航天大学 | A kind of segmentation of examinee's posture and smoothing method |
| CN110321754A (en) * | 2018-03-28 | 2019-10-11 | 西安铭宇信息科技有限公司 | A kind of human motion posture correcting method based on computer vision and system |
| CN111144217A (en) * | 2019-11-28 | 2020-05-12 | 重庆邮电大学 | Motion evaluation method based on human body three-dimensional joint point detection |
| CN112819852A (en) * | 2019-11-15 | 2021-05-18 | 微软技术许可有限责任公司 | Evaluating gesture-based motion |
-
2021
- 2021-06-04 CN CN202110626797.6A patent/CN113392745A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110321754A (en) * | 2018-03-28 | 2019-10-11 | 西安铭宇信息科技有限公司 | A kind of human motion posture correcting method based on computer vision and system |
| CN110163112A (en) * | 2019-04-25 | 2019-08-23 | 沈阳航空航天大学 | A kind of segmentation of examinee's posture and smoothing method |
| CN112819852A (en) * | 2019-11-15 | 2021-05-18 | 微软技术许可有限责任公司 | Evaluating gesture-based motion |
| CN111144217A (en) * | 2019-11-28 | 2020-05-12 | 重庆邮电大学 | Motion evaluation method based on human body three-dimensional joint point detection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11741367B2 (en) | Apparatus and method for image processing to calculate likelihood of image of target object detected from input image | |
| CN108205654B (en) | Action detection method and device based on video | |
| CN109448090B (en) | Image processing method, device, electronic equipment and storage medium | |
| CN104268591B (en) | A kind of facial critical point detection method and device | |
| US11423699B2 (en) | Action recognition method and apparatus and electronic equipment | |
| CN109101901B (en) | Human body action recognition method and device, neural network generation method and device and electronic equipment | |
| CN111062239A (en) | Human body target detection method and device, computer equipment and storage medium | |
| JP2013065103A (en) | Tracking apparatus, tracking method and program | |
| US20150117753A1 (en) | Computing device and method for debugging computerized numerical control machine | |
| CN112633196A (en) | Human body posture detection method and device and computer equipment | |
| CN107563295B (en) | Multi-Kinect-based all-dimensional human body tracking method and processing equipment | |
| CN111696130A (en) | Target tracking method, target tracking apparatus, and computer-readable storage medium | |
| CN104657709A (en) | Face image identification method and device as well as server | |
| CN110717476A (en) | Image processing method, image processing device, computer equipment and computer readable storage medium | |
| CN110598647B (en) | Head posture recognition method based on image recognition | |
| CN110991292A (en) | Action identification comparison method and system, computer storage medium and electronic device | |
| CN110910426A (en) | Action process and action trend identification method, storage medium and electronic device | |
| CN112699784A (en) | Face orientation estimation method and device, electronic equipment and storage medium | |
| CN110956131B (en) | Single-target tracking method, device and system | |
| CN113392745A (en) | Abnormal action correction method, abnormal action correction device, electronic equipment and computer storage medium | |
| CN112509154A (en) | Training method of image generation model, image generation method and device | |
| US10013602B2 (en) | Feature vector extraction device based on cell positioning | |
| US11527090B2 (en) | Information processing apparatus, control method, and non-transitory storage medium | |
| CN113392743B (en) | Abnormal action detection method, abnormal action detection device, electronic equipment and computer storage medium | |
| CN112307799A (en) | Gesture recognition method, device, system, storage medium and equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210914 |
|
| RJ01 | Rejection of invention patent application after publication |