Motion recognition method and device, electronic equipment and storage medium

Abstract

The invention discloses a motion recognition method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a target human body motion image corresponding to a target user, and performing limb recognition on the target human body motion image to obtain target limb recognition data corresponding to the target user; and finally, based on the target physiological parameter data and the target limb identification data in the motion process of the target user, performing motion evaluation on the target user to obtain a target motion evaluation result, the target motion evaluation result comprising the current motion state of the target user, and the target user is in the motion state. Therefore, the target limb recognition data and the target physiological parameter data of the target user are comprehensively considered, comprehensive and accurate evaluation of the motion condition of the target user in the motion process is achieved, the accuracy and scientificity of motion recognition are improved, and real-time motion feedback can be provided for the user.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a motion recognition method, device, electronic device, and storage medium. Background Art

[0002] With the rapid development of network technology, network security issues are becoming increasingly prominent, and various cyberattack methods are constantly emerging, posing significant challenges to network security. Network traffic data, as a key threat intelligence carrier that records network behavior, may contain a variety of potential attacks. Therefore, how to effectively detect attacks from this threat intelligence has become a crucial step in improving network security protection capabilities. Summary of the Invention

[0003] The embodiments of the present application provide a motion recognition method, device, electronic device, and storage medium that comprehensively consider the target user's target limb recognition data and target physiological parameter data, thereby achieving a comprehensive and accurate assessment of the target user's motion during exercise, improving the accuracy and scientific nature of motion recognition, and providing real-time motion feedback to the user. The above technical solution is as follows:

[0004] In a first aspect, an embodiment of the present application provides a motion recognition method, the method comprising:

[0005] Obtaining a target human motion image corresponding to a target user;

[0006] Performing limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user;

[0007] Based on the target physiological parameter data and target limb recognition data of the target user during exercise, a motion evaluation is performed on the target user to obtain a target motion evaluation result; the target motion evaluation result includes the current motion state of the target user.

[0008] In a possible implementation, performing limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user includes:

[0009] Based on the target motion environment data corresponding to the target user, limb recognition is performed on the target human motion image to obtain target limb recognition data corresponding to the target user.

[0010] In one possible implementation, the target limb recognition data includes a sequence of first limb recognition results within a current time period; the sequence of first limb recognition results is composed of first limb recognition results corresponding to multiple frames of target human motion images acquired within the current time period, arranged in chronological order of acquisition; and the target physiological parameter data includes a sequence of first physiological parameters of the target user's current exercise within the current time period.

[0011] The target user is evaluated for movement based on the target physiological parameter data and target limb recognition data during movement, and the target movement evaluation result is obtained, including:

[0012] determining the current exercise state of the target user based on the first limb recognition result sequence and the first physiological parameter sequence, and determining the current exercise intensity information of the target user based on the first limb recognition result sequence;

[0013] matching the first limb recognition result sequence with a preset limb motion data sequence corresponding to at least one preset motion action, and determining the current motion action of the target user within the current time period based on the corresponding limb motion matching results;

[0014] Comparing the current exercise intensity information corresponding to the current exercise action with the preset exercise intensity information, and determining the current action standard evaluation result corresponding to the current exercise action based on the corresponding exercise intensity comparison result;

[0015] determining a current sport of the target user based on a target sport action sequence of the target user during the current sport;

[0016] The current exercise quality evaluation result corresponding to the current exercise project is determined based on the target exercise intensity information corresponding to each target exercise action in the current exercise project and the preset exercise intensity information.

[0017] In a possible implementation, determining the current motion state of the target user based on the first limb recognition result sequence and the first physiological parameter sequence includes:

[0018] Determining current physiological parameter change information of the target user based on the first physiological parameter sequence, the preset physiological parameter threshold corresponding to the target user, and / or the second physiological parameter sequence within the initial time period of the target user's current exercise;

[0019] Determining current limb movement change information of the target user based on the first limb recognition result sequence, the preset limb movement data sequence of the target user corresponding to the current movement action, and / or the second limb recognition result sequence within the initial time period;

[0020] The current motion state of the target user is determined based on the current physiological parameter change information and / or the current limb movement change information.

[0021] In one possible implementation, before performing a motion assessment on the target user based on the target physiological parameter data and the target limb recognition data during the target user's motion, and obtaining the target motion assessment result, the method further includes:

[0022] Using environmental sensors to obtain target motion environment data corresponding to the target user;

[0023] In a case where the target motion environment data exceeds a preset motion environment data range, the preset motion intensity information corresponding to the current motion action is adjusted based on target difference information between the target motion environment data and the preset motion environment data range.

[0024] In a possible implementation, the target motion evaluation result further includes at least one of the following: a current action standard evaluation result and a current motion quality evaluation result corresponding to the target user;

[0025] After performing a motion assessment on the target user based on the target physiological parameter data and target limb recognition data during the target user's motion and obtaining a target motion assessment result, the method further includes:

[0026] Generate and issue current action guidance information corresponding to the target user based on the current action standard evaluation result; and / or

[0027] Generate and send current exercise quality reminder information corresponding to the target user based on the current exercise quality assessment result; and / or

[0028] In the case that the current exercise program of the target user is not completed, if the current exercise state is the stopped exercise state and the current duration corresponding to the stopped exercise state exceeds the target duration, the screaming mode is activated to emit a target warning sound, and after the current exercise state changes from the stopped exercise state to the exercising state, the screaming mode is turned off to stop emitting the target warning sound; the target emission volume corresponding to the target warning sound is greater than or equal to the target value; and / or

[0029] When the current exercise state is a fatigue state, a target encouragement message is sent.

[0030] In a possible implementation, the method further includes:

[0031] Receive a target warning sound configuration operation input by the target user; the target warning sound configuration operation carries target configuration parameters selected by the target user for the target warning sound; the target configuration parameters include at least one of the following: target type, target volume, target duration, target trigger frequency, and target trigger condition corresponding to the target warning sound;

[0032] In response to the target warning sound configuration operation, the target warning sound corresponding to the target user's exercise process is configured based on the target configuration parameters.

[0033] In one possible implementation, before initiating the screaming mode to emit the target warning sound, the method further includes:

[0034] Obtaining the target voice command issued by the target user;

[0035] In response to the above-mentioned target voice command, the screaming mode when not moving is turned on; the screaming mode when not moving is used to start the screaming mode when the above-mentioned target user is in a stopped state and the current duration corresponding to the stopped state exceeds the above-mentioned target duration.

[0036] In a possible implementation, before performing limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user, the method further includes:

[0037] receiving a designated exercise type inputted by the target user;

[0038] The above-mentioned body recognition is performed on the above-mentioned target human motion image to obtain the target body recognition data corresponding to the above-mentioned target user, including:

[0039] Using a specified limb recognition algorithm or a specified limb recognition model corresponding to the specified motion type, performing limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user;

[0040] The above-mentioned different designated motion types correspond to different limb motion parameters captured during the limb recognition process.

[0041] In a second aspect, an embodiment of the present application provides a motion recognition device, comprising:

[0042] An acquisition module, used to acquire a target human motion image corresponding to a target user;

[0043] A limb recognition module is used to perform limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user;

[0044] The motion evaluation module is used to perform motion evaluation on the target user based on the target physiological parameter data and target limb recognition data during the target user's motion process to obtain a target motion evaluation result; the target motion evaluation result includes the current motion state of the target user.

[0045] In a third aspect, an embodiment of the present application provides an electronic device, including: a processor and a memory;

[0046] The processor is connected to the memory;

[0047] The aforementioned memory is used to store executable program code;

[0048] The processor runs the program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method provided by the first aspect of the embodiment of this specification or any possible implementation of the first aspect.

[0049] In a fourth aspect, an embodiment of this specification provides a computer storage medium, which stores multiple instructions, and the instructions are suitable for being loaded by a processor and executing the method provided by the first aspect of the embodiment of this specification or any possible implementation of the first aspect.

[0050] In an embodiment of the present application, a target human body motion image corresponding to the target user is obtained, and limb recognition is performed on the above-mentioned target human body motion image to obtain target limb recognition data corresponding to the above-mentioned target user; finally, based on the target physiological parameter data and target limb recognition data of the above-mentioned target user during the movement process, the above-mentioned target user is evaluated for movement to obtain a target movement evaluation result. The above-mentioned target movement evaluation result includes the current movement state of the above-mentioned target user, thereby comprehensively considering the target limb recognition data and target physiological parameter data of the target user, realizing a comprehensive and accurate evaluation of the movement situation of the target user during the movement process, improving the accuracy and scientificity of movement recognition, and being able to provide real-time movement feedback to the user.

[0051] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments. 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.

[0053] Figure 1 A schematic diagram of the architecture of a motion recognition system provided by an exemplary embodiment of the present application;

[0054] Figure 2 A flowchart of a motion recognition method provided by an exemplary embodiment of the present application;

[0055] Figure 3 A schematic diagram of a motion evaluation implementation process provided by an exemplary embodiment of the present application;

[0056] Figure 4 A schematic diagram of a process for determining a current motion state provided by an exemplary embodiment of the present application;

[0057] Figure 5 A flowchart of another motion recognition method provided by an exemplary embodiment of the present application;

[0058] Figure 6 This is a flowchart of another motion recognition method according to an exemplary embodiment of the present application;

[0059] Figure 7 A schematic structural diagram of a motion recognition device provided by an exemplary embodiment of the present application;

[0060] Figure 8 A schematic structural diagram of an electronic device provided as an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0061] To make the features and advantages of this application more obvious and easy to understand, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of this application.

[0062] The terms "first," "second," "third," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish between different objects, not to describe a particular order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements, but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0063] Please refer to the following Figure 1 , which is a schematic diagram of the architecture of a motion recognition system provided by an exemplary embodiment of this specification. Figure 1 As shown, the motion recognition system may include: an electronic device 110 and a server 120 .

[0064] in:

[0065] The electronic device 110 may be one or more image acquisition devices, or a terminal equipped with an image acquisition device or connected to one or more image acquisition devices. A user version of software may be installed in the electronic device 110 to implement corresponding motion recognition during the user's movement. The electronic device 110 may obtain a target human motion image corresponding to the target user, and perform limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user; finally, based on the target physiological parameter data and target limb recognition data of the target user during the movement, the target user is subjected to motion evaluation to obtain a target motion evaluation result. The target motion evaluation result includes the current motion state of the target user.

[0066] Optionally, after obtaining the target human motion image corresponding to the target user, the electronic device 110 can also, but is not limited to, send the target human motion image corresponding to the above-mentioned target user to the server 120 via the network, so that the server 120 can subsequently realize corresponding motion recognition of the target user through the improved motion recognition method of the embodiment of the present application, so as to assist the target user to exercise better.

[0067] Server 120 can be a server that provides multiple motion recognition services. It can obtain the target human motion image sent by electronic device 110 via the network, perform limb recognition on the target human motion image, obtain target limb recognition data corresponding to the target user, and perform motion assessment on the target user based on the target physiological parameter data and target limb recognition data during the target user's motion to obtain a target motion assessment result. Server 120 can be, but is not limited to, a hardware server, a virtual server, a cloud server, etc.

[0068] It can be understood that the motion recognition method provided in the embodiment of the present application can be executed by the electronic device 110 or the server 120 alone, or can be executed by the electronic device 110 and the server 120 together, and the embodiment of the present application is not limited to this.

[0069] The network may be a medium that provides a communication link between any electronic device 110 and the server 120, or may be the Internet including network devices and transmission media, but is not limited thereto. The transmission media may be a wired link, such as, but not limited to, a coaxial cable, optical fiber, and a digital subscriber line (DSL), or a wireless link, such as, but not limited to, wireless fidelity (WIFI), Bluetooth, and a mobile device network.

[0070] Understandably, Figure 1 The number of electronic devices 110 and servers 120 in the motion recognition system shown is for illustrative purposes only. In a specific implementation, the motion recognition system may include any number of electronic devices 110 and servers 120. This specification does not specifically limit this. For example, but not limited to, the electronic device 110 may be an electronic device cluster consisting of multiple electronic devices, and the server 120 may be a server cluster consisting of multiple servers.

[0071] Please refer to the following Figure 2 , taking the electronic device performing motion recognition as an example, a motion recognition method provided by an exemplary embodiment of the present application is introduced. Figure 2 As shown, the motion recognition method includes the following steps:

[0072] S201, obtaining a target human motion image corresponding to a target user.

[0073] Specifically, the target human motion image is an image collected during the target user's motion, and may include, but is not limited to, key limb information such as the overall outline and limb movements of the target user during the motion.

[0074] Optionally, before exercising, the target user can turn on the image acquisition device installed on his terminal, so that during the target user's exercise, the target human body motion image of the target user can be captured in real time through the image acquisition device installed on the terminal, and real-time motion recognition of the target user can be realized based on the captured target human body motion image.

[0075] Optionally, before exercising, the target user may also move directly into the visual detection range of an image acquisition device with motion recognition function, so that during the target user's exercise, the image acquisition device can capture the target human body motion image corresponding to the target user within its visual detection range in real time.

[0076] Optionally, after capturing the original human motion image of the target user during movement, the electronic device may, but is not limited to, pre-process the captured original human motion image, such as but not limited to denoising processing (to eliminate noise points in the image and improve image quality), contrast enhancement processing (to make key information in the image more prominent to facilitate subsequent feature extraction), image cropping and scaling processing, etc., to obtain the target human motion image, thereby ensuring the quality of the target human motion image and improving the efficiency and accuracy of subsequent limb recognition based on the target human motion image.

[0077] S202 , performing limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user.

[0078] Specifically, after acquiring the target human motion image, a limb recognition algorithm or a pre-trained limb recognition model can be used to identify target limb recognition data, such as the position information of key limb nodes (such as, but not limited to, shoulder joints, elbow joints, knee joints, etc.) corresponding to the target user in the target human motion image, as well as the relative position information between key limb nodes. The limb recognition model can be, but is not limited to, trained based on multiple human motion images with known limb recognition data.

[0079] Optionally, the above-mentioned S202, the implementation process of performing limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user, may include, but is not limited to: performing limb recognition on the target human motion image based on the target motion environment data corresponding to the target user to obtain target limb recognition data corresponding to the target user. For example, but not limited to, the parameters of the limb recognition algorithm may be automatically adjusted based on the target motion environment data, and then the adjusted limb recognition algorithm may be used to perform limb recognition on the target human motion image to obtain the corresponding target limb recognition data; or the target motion environment data and the target human motion image may be directly input into a limb recognition model, and the limb recognition model may perform limb recognition on the target human motion image in combination with the target motion environment data, and output the corresponding target limb recognition data. The above-mentioned limb recognition model may be, but is not limited to, trained based on multiple human motion images with known limb recognition data and the motion environment data corresponding to each human motion image.

[0080] The above-mentioned target motion environment data may include, but is not limited to, at least one of the following: light intensity, rain / snow / fog concentration, image acquisition distance, background congestion, computing resource data, device jitter data, and temperature. The above-mentioned light intensity may be detected by a light sensor installed on the electronic device, or may be calculated based on the pixel brightness or grayscale value of the target human area corresponding to the target user in the target human motion image, and the embodiments of the present application are not limited thereto. The above-mentioned rain / snow / fog concentration may be, but is not limited to, obtained from a weather station via the network. The above-mentioned image acquisition distance is the distance between the target user and the image acquisition device or the image acquisition device installed on the electronic device. The above-mentioned background congestion may be, but is not limited to, characterizing the number of human bodies contained in the target human motion image. The above-mentioned computing resource data may include, but is not limited to, the current load of the electronic device. The above-mentioned device jitter data is used to characterize the jitter of the device that captures the target human motion image, and may be, but is not limited to, detected by a gyroscope.

[0081] For example, when the perceived light intensity is lower than a preset "low-light threshold", or the average brightness or average contrast of the target human motion image is lower than the target value, it can be determined that the current environment is a low-light environment, thereby triggering the parameter adaptive adjustment module of the limb recognition algorithm to make corresponding adjustments to the low threshold and high threshold of the edge detection operator in the limb recognition algorithm, the key point confidence threshold, the filtering intensity, the contrast stretching amplitude and other parameters, thereby improving the visibility and robustness of the limb features, while suppressing noise interference and improving the recognition accuracy.

[0082] In the embodiment of the present application, different from the traditional single limb recognition or motion monitoring technology, the embodiment of the present application adopts a multimodal fusion analysis method to comprehensively analyze the target human motion image and the target motion environment data to more comprehensively and accurately identify the target limb recognition data of the target user.

[0083] S203 , performing a motion evaluation on the target user based on the target physiological parameter data and the target limb recognition data during the target user's motion process to obtain a target motion evaluation result.

[0084] Specifically, the target physiological parameter data can be obtained by detecting the target physiological parameters of the target user during the movement of the target user by a smart wearable device that is communicatively connected to the electronic device, or by analyzing and evaluating the target physiological parameters of the target user during the movement of the target human body motion image. The embodiment of the present application does not limit this. The target physiological parameter data can include, but is not limited to, at least one of the following: target heart rate, respiratory rate, blood oxygen saturation, blood pressure, oxygen uptake, and body temperature during the movement of the target user. The target limb recognition data includes target limb feature information of the target user identified in the target human body motion image, such as, but not limited to, position information of the target user's corresponding limb key nodes (such as, but not limited to, shoulder joints, elbow joints, knee joints, etc.) and relative position information between each limb key node, limb movement type, limb movement trajectory, etc. The target motion evaluation result can include, but is not limited to, the current motion state of the target user.

[0085] In an embodiment of the present application, a target human body motion image corresponding to the target user is obtained, and limb recognition is performed on the above-mentioned target human body motion image to obtain target limb recognition data corresponding to the above-mentioned target user; finally, based on the target physiological parameter data and target limb recognition data of the above-mentioned target user during the movement process, the above-mentioned target user is evaluated for movement to obtain a target movement evaluation result. The above-mentioned target movement evaluation result includes the current movement state of the above-mentioned target user, thereby comprehensively considering the target limb recognition data and target physiological parameter data of the target user, realizing a comprehensive and accurate evaluation of the movement situation of the target user during the movement process, improving the accuracy and scientificity of movement recognition, and being able to provide real-time movement feedback to the user.

[0086] Optionally, after obtaining the target limb recognition data and target physiological parameter data of the target user during movement, the target limb recognition data and target physiological parameter data can be input into a motion assessment model to extract the target limb features corresponding to the target limb recognition data and the target physiological features corresponding to the target physiological parameter data. The target limb features and the target physiological features are then fused to obtain a target fusion feature. Finally, based on the target fusion feature, a corresponding target motion assessment result is output. The above-mentioned motion assessment model can be, but is not limited to, trained based on multiple limb recognition data with known motion assessment results and the physiological parameter data corresponding to each limb recognition data.

[0087] Optionally, after obtaining the target limb recognition data, target physiological parameter data, and target motion environment data during the target user's motion, the target limb recognition data, target physiological parameter data, and target motion environment data can be input into a motion evaluation model to extract the target limb features corresponding to the target limb recognition data, the target physiological features corresponding to the target physiological parameter data, and the target motion environment features corresponding to the target motion environment data. The target limb features, target physiological features, and target motion environment features are then fused to obtain a target fusion feature, and finally, the corresponding target motion evaluation result is output based on the target fusion feature. The above-mentioned motion evaluation model can be trained, but is not limited to, based on multiple limb recognition data with known motion evaluation results and the physiological parameter data and motion environment data corresponding to each limb recognition data.

[0088] In the embodiment of the present application, different from the traditional single limb recognition or motion monitoring technology, the embodiment of the present application adopts a multimodal fusion analysis method to comprehensively analyze the target limb recognition data, target physiological parameter data and target motion environment data during the target user's movement process to more comprehensively and accurately evaluate the target user's motion status.

[0089] In some possible embodiments, the target limb recognition data may include, but is not limited to, a first limb recognition result sequence within the current time period, and the first limb recognition result sequence is composed of the first limb recognition results corresponding to multiple frames of target human motion images acquired within the current time period, arranged in chronological order of acquisition time. The target physiological parameter data may include, but is not limited to, a first physiological parameter sequence within the current time period of the target user's current exercise, and the first physiological parameter sequence is composed of multiple first physiological parameter data of the target user within the current time period, arranged in chronological order of acquisition time. The current time period may change with the current moment, that is, the time interval corresponding to the current time period remains unchanged, and its end moment is the current moment. Figure 3 As shown, the above S203, based on the target physiological parameter data and target limb recognition data of the target user during the movement, performs a movement evaluation on the target user, and the implementation process of obtaining the target movement evaluation result may include but is not limited to:

[0090] S301 : Determine a current motion state of a target user based on a first limb recognition result sequence and a first physiological parameter sequence, and determine current motion intensity information of the target user based on the first limb recognition result sequence.

[0091] Specifically, the current exercise state is used to characterize the individual's immediate physiological and behavioral performance during exercise. The current exercise intensity information may include, but is not limited to, information related to exercise intensity, such as current exercise amplitude and / or current exercise frequency.

[0092] Optionally, it is possible but not limited to first determining first limb change information (such as but not limited to limb movement trajectory, limb change amplitude, limb change frequency, etc.) within the current time period based on the first limb recognition result sequence and determining first physiological parameter information (such as but not limited to heart rate change amplitude, average heart rate, respiratory rate change amplitude, maximum respiratory rate, maximum heart rate, etc.) within the current time period based on the first physiological parameter sequence, and then determining the current motion state of the target user based on the above-mentioned first limb change information and the first physiological parameter information according to the preset limb change information and the correspondence between the physiological parameter information and the motion state. For example, but not limited to, when the limb change amplitude and / or limb change frequency of the target user within the current time period in the first limb change information is less than a first threshold and the heart rate change amplitude in the first physiological parameter information is less than the first amplitude, it is determined that the current motion state of the target user is a non-motion state.

[0093] It can be understood that the first limb change information and / or the first physiological parameter information belong to different interval ranges, corresponding to different current states of the target user.

[0094] Alternatively, as Figure 4 As shown, the implementation process of determining the current motion state of the target user based on the first limb recognition result sequence and the first physiological parameter sequence in the above S301 may include, but is not limited to:

[0095] S401 : Determine current physiological parameter change information of the target user based on a first physiological parameter sequence, a preset physiological parameter threshold corresponding to the target user, and / or a second physiological parameter sequence within an initial time period of the target user's current exercise.

[0096] Specifically, the time interval corresponding to the above-mentioned initial time period is the same as the time interval corresponding to the current time period. The above-mentioned initial time period can be the starting time period of the target user's current exercise, or it can be the previous time period before the current time period. It can be set specifically according to the actual exercise evaluation needs, and the embodiments of the present application do not limit this. The above-mentioned preset physiological parameter threshold can be the maximum physiological parameter value preset by the target user, or it can be the normal physiological parameter value when the target user is in a non-exercise state, etc. It can be set specifically according to actual needs, and the embodiments of the present application do not limit this. The current physiological parameter change information of the target user can be determined based on, but not limited to, the difference information between the first physiological parameter sequence and the preset physiological parameter threshold corresponding to the target user and / or the difference information between the first physiological parameter sequence and the second physiological parameter sequence within the initial time period of the target user's current exercise.

[0097] S402 : Determine current limb movement change information of the target user based on the first limb recognition result sequence, the preset limb movement data sequence of the target user corresponding to the current movement action, and / or the second limb recognition result sequence within the initial time period.

[0098] Specifically, the above-mentioned current motion action can be determined by the method described in S203 below, or can be obtained by input from the target user, which is not limited in this embodiment of the present application. The time interval of the time period corresponding to the above-mentioned preset limb motion data sequence is equal to the time interval of the current time period. The above-mentioned current limb motion change information can be, but is not limited to, based on the difference information between the first limb recognition result sequence and the preset limb motion data sequence corresponding to the current motion action of the target user, and / or, the difference information between the first limb recognition result sequence and the second limb recognition result sequence within the initial time period.

[0099] S403: Determine the current exercise state of the target user based on the current physiological parameter change information and / or the current limb movement change information.

[0100] Optionally, the current physiological parameter change information and / or current limb movement change information may be input into a motion state assessment model, but is not limited to inputting the current motion state of the target user. The motion state assessment model may be a multimodal temporal neural network model, which may be, but is not limited to, fusing the current physiological parameter change information and the current limb movement change information to obtain a current change fusion feature, and outputting the corresponding current motion state based on the current change fusion feature.

[0101] Optionally, the current motion state of the target user can also be determined based on, but not limited to, current physiological parameter change information and / or current limb movement change information, according to the correspondence between each preset physiological parameter change range and / or each preset limb movement change range and each motion state.

[0102] For example, if the target user's heart rate in the current time period reaches 80% of the preset maximum heart rate, and the current physiological parameter change information indicates that the breathing rate has significantly accelerated, which is 30% higher than the breathing rate in the initial time period, and at the same time, the current limb movement change information indicates that the target user's current exercise action is slow, for example but not limited to, the jumping height during rope skipping (current exercise action) is 20% lower than the jumping height in the initial time period, or the getting-up speed during burpees (current exercise action) is 15% slower than the getting-up speed in the initial time period, etc., it can be determined that the target user is currently in a state of exercise fatigue.

[0103] Please continue to refer to Figure 3 ,like Figure 3 As shown, in the above S301, after determining the current exercise state of the target user based on the first limb recognition result sequence and the first physiological parameter sequence, and determining the current exercise intensity information of the target user based on the first limb recognition result sequence, the implementation process of the above S203 may also include, but is not limited to:

[0104] S302: Match the first limb recognition result sequence with a preset limb motion data sequence corresponding to at least one preset motion action, and determine the current motion of the target user in the current time period based on the corresponding limb motion matching results.

[0105] Specifically, the limb motion matching result includes the degree of match (such as, but not limited to, the degree of similarity in limb motion trajectories) between the first limb recognition result sequence and the preset limb motion data sequence corresponding to each preset motion. The preset motion with the highest degree of match or a degree greater than a target degree of match (such as, but not limited to, 95%, 90%, etc.) may be determined as the current motion of the target user.

[0106] S303: Compare the current exercise intensity information corresponding to the current exercise action with the preset exercise intensity information, and determine the current action standard evaluation result corresponding to the current exercise action based on the corresponding exercise intensity comparison result.

[0107] Specifically, the preset exercise intensity information may include, but is not limited to, information related to exercise intensity, such as a preset exercise amplitude and / or preset exercise frequency corresponding to the current exercise action. The exercise intensity comparison result may include, but is not limited to, an exercise intensity difference between the current exercise intensity information and the preset exercise intensity information. When the exercise intensity difference is greater than a target difference, it can be determined that the current exercise action is not performed accurately, i.e., not in accordance with the standard. When the exercise intensity difference is less than or equal to the target difference, it can be determined that the current exercise action is performed accurately, i.e., in accordance with the standard.

[0108] Optionally, the current action standard evaluation result corresponding to the above-mentioned current motion action may also be determined based on, but not limited to, the corresponding limb motion matching result obtained in S302, or may be determined based on the corresponding limb motion matching result obtained in S302 and the motion intensity comparison result obtained in S303. That is, the limb motion matching result obtained in S302 may also include, but not limited to, limb motion difference information (such as, but not limited to, limb motion amplitude difference, limb motion trajectory difference, etc.) between the first limb recognition result sequence and the preset limb motion data sequence corresponding to the current motion action. The current action standard evaluation result corresponding to the current motion action can be directly determined based on the above-mentioned limb motion difference information. For example, but not limited to, when the limb motion amplitude difference or the limb motion trajectory difference is greater than a certain value, it is determined that the current action standard evaluation result is non-standard for the current motion action.

[0109] S304: Determine the current sport of the target user based on the target sport action sequence during the current sport of the target user.

[0110] Specifically, the target motion sequence is composed of at least one target motion identified during the target user's current motion, arranged in a sequence of motions. Because different sports events contain different motions and / or require different sequences of motion for each motion, the target motion sequence during the target user's current motion can be compared with preset motion sequences corresponding to each preset sports event to determine the target user's current sports event. Alternatively, the target motion sequence during the target user's current motion can be directly input into a pre-trained sports event evaluation model or a target macro model, with the model outputting the corresponding current sports event. This embodiment of the present application does not limit this.

[0111] In the embodiments of the present application, by detecting and tracking human joints and matching body movements, the target user's current sport can be accurately determined, and the accuracy and standardization of the target user's movements can be evaluated. For example, when the target user is identified as skipping rope, the target user's skipping frequency, jump height, and arm swing amplitude can be accurately determined. When the target user is identified as doing push-ups, key movement standard indicators such as the target user's arm flexion and extension angle and whether the body maintains a straight line can be detected.

[0112] S305 : Determine a current exercise quality evaluation result corresponding to the current exercise program based on target exercise intensity information corresponding to each target exercise action in the current exercise program and preset exercise intensity information.

[0113] Specifically, the target exercise intensity information is the actual exercise intensity information of the target user performing the corresponding target exercise action during the current exercise. If the target user has not yet performed the target exercise action, the target exercise intensity information corresponding to the target exercise action is empty. The target exercise intensity corresponding to the target exercise action currently being performed (i.e., the current exercise action) is updated as the target user's current exercise status is updated. The current exercise quality assessment results may include, but are not limited to, the current completion progress of the current exercise event and the current completion quality score.

[0114] Optionally, after the target user finishes the exercise, the target action standard evaluation results corresponding to each target motion action can be determined in the manner described in S303 above, and then the current motion quality evaluation results corresponding to the current motion item can be determined based on the target action standard evaluation results corresponding to each target motion action.

[0115] Optionally, during exercise, the target user may, but is not limited to, evaluate the current exercise quality corresponding to the current exercise project based on the target exercise intensity information corresponding to the target exercise action completed by the target user in the current exercise project and the target exercise intensity information corresponding to the target exercise action currently in progress.

[0116] In some possible embodiments, before performing a motion assessment on the target user based on the target physiological parameter data and target limb recognition data during the target user's motion process and obtaining the target motion assessment result, the target motion environment data corresponding to the target user can also be obtained using an environmental sensor; then, when the target motion environment data exceeds the preset motion environment data range, the preset motion intensity information corresponding to the current motion action is adjusted based on the target difference information between the target motion environment data and the preset motion environment data range (for example, but not limited to, based on the above-mentioned target difference information, the preset motion intensity information is adjusted according to the correspondence between the preset motion environment difference information and the motion intensity adjustment information). For example, but not limited to, when the temperature is high, the requirements for the target user's motion intensity can be appropriately reduced according to the temperature difference. At the same time, it is also possible, but not limited to, to remind the target user to replenish water, so as to provide the user with more personalized and humane exercise guidance and supervision services.

[0117] It can be understood that the above-mentioned preset exercise intensity information can be general exercise intensity information corresponding to the exercise action, or it can be exercise intensity information pre-entered by the target user in the electronic device, or it can be exercise intensity information automatically generated based on the target user information and / or the target user's exercise goals (such as but not limited to weight loss, body shaping, etc.) using the target macro model, that is, exercise intensity information customized to suit the target user's physical fitness according to their needs, which can be set specifically according to the actual exercise needs of the target user, and the embodiments of the present application are not limited to this.

[0118] Please refer to the following Figure 5 , which is a flow chart of another motion recognition method provided by an exemplary embodiment of the present application. Figure 5 As shown, the motion recognition method may include but is not limited to the following steps:

[0119] S501: Receive a target warning sound configuration operation input by a target user, where the target warning sound configuration operation carries target configuration parameters selected by the target user for the target warning sound.

[0120] Specifically, the electronic device has a motion alert function (i.e., a screaming mode when not exercising). When a target user wishes to configure a target alert sound corresponding to the motion alert function, the target user can directly input the corresponding target alert sound configuration operation via the electronic device's display and / or keypad. The target configuration parameters may include, but are not limited to, at least one of the following: a target type corresponding to the target alert sound, a target volume, a target duration, a target trigger frequency, and a target trigger condition.

[0121] S502 : In response to the target warning sound configuration operation, configure a target warning sound corresponding to the target user's exercise process based on target configuration parameters.

[0122] Specifically, after receiving the target user's input for configuring the target warning sound, the electronic device can directly configure the corresponding target warning sound during the target user's exercise based on the target configuration parameters, thereby providing the target user with a personalized "screaming if you don't exercise" mode configuration function. The target user can choose different types of target warning sounds according to their preferences, such as sharp alarms, exciting music, or custom recorded sounds. At the same time, the target user can also set parameters such as the volume, duration, and trigger frequency of the target warning sound. For example, in a home environment, the target user may want the target warning sound volume to be moderate to avoid disturbing others; while in a personal fitness space, the target user can turn up the volume to enhance the supervision effect, meeting the diverse needs of different users.

[0123] In some possible embodiments, such as Figure 5 As shown, the motion recognition method may also include but is not limited to:

[0124] S503: Acquire a target voice command issued by a target user.

[0125] Specifically, the electronic device is equipped with a highly sensitive voice recognition module that can accurately identify the target user's voice command to activate the "No Exercise, Scream Mode" (i.e., the target voice command). This voice recognition module has been trained with a large amount of voice data and has strong noise immunity. It can accurately capture the target user's voice in a variety of environments, ensuring the convenience of activating the motion alert function (i.e., No Exercise, Scream Mode).

[0126] S504: In response to the target voice command, the screaming mode is turned on without exercising.

[0127] Specifically, the above-mentioned screaming mode when not moving is used to activate the screaming mode when the target user is in a stopped state and the current duration corresponding to the stopped state exceeds the target duration (for example, but not limited to 1 minute, 30 seconds, 2 minutes, etc.).

[0128] S505: Acquire a target human motion image corresponding to the target user.

[0129] Specifically, the above S505 is consistent with the above S201 and will not be repeated here.

[0130] S506: Perform limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user.

[0131] Specifically, the above S506 is consistent with the above S202 and will not be repeated here.

[0132] S507 , performing a motion evaluation on the target user based on the target physiological parameter data and the target limb recognition data during the target user's motion process to obtain a target motion evaluation result.

[0133] Specifically, the above S507 is consistent with the above S203 and will not be repeated here.

[0134] In some possible embodiments, the target motion evaluation result may also include, but is not limited to, at least one of the following: a current action standard evaluation result corresponding to the target user, and a current motion quality evaluation result. Figure 5 As shown, after performing a motion evaluation on the target user based on the target physiological parameter data and the target limb recognition data during the target user's motion process to obtain a target motion evaluation result, the motion recognition method may further include, but is not limited to:

[0135] S508: Generate and issue current action guidance information corresponding to the target user based on the current action standard evaluation result.

[0136] Specifically, when the current action standard evaluation result indicates that the target user's current motion action is not standard, the current action guidance information may be, but is not limited to, generated based on the correct action description information of the current motion action and / or the factor information that causes the target user's current motion action to be non-standard based on the target physiological parameter data and target limb recognition data during the target user's exercise. When the current action standard evaluation result indicates that the target user's current motion action is standard, the current action guidance information may be, but is not limited to, including the correct action description information of the current motion action.

[0137] In the related art, only motion can be recorded or simple motion recognition can be performed, while the electronic device in the embodiment of the present application has a real-time voice action correction guidance function. During the target user's motion process, once it is detected that its action is inappropriate, that is, the current action standard evaluation result indicates that the current motion of the target user is not standard, it will immediately be described in detail by clear voice the correct action essentials of the current motion (i.e., current action guidance information). For example, when the target user does burpees, if it is detected that the target user gets up too quickly and causes the body to shake too much, the electronic device will send a voice prompt "please keep your body stable when getting up, get up slowly, and control your body balance" (current action guidance information), help the target user quickly understand and correct errors, and greatly improve the effect and user experience of motion guidance.

[0138] and / or

[0139] S509: Generate and send current exercise quality reminder information corresponding to the target user based on the current exercise quality evaluation result.

[0140] Specifically, the current exercise quality assessment result may include, but is not limited to, exercise effect data and / or an exercise quality score achieved by the target user during the current exercise, such as calories burned. The current exercise quality reminder information is used to inform the target user of the current exercise quality to motivate the target user to continue exercising.

[0141] and / or

[0142] S510, when the current exercise program of the target user is not completed, if the current exercise state is the stop exercise state and the current duration corresponding to the stop exercise state exceeds the target duration, the screaming mode is activated to emit the target warning sound, and after the current exercise state changes from the stop exercise state to the exercise state, the screaming mode is turned off to stop emitting the target warning sound.

[0143] Specifically, the target sound volume corresponding to the target warning sound is greater than or equal to the target value.

[0144] For example, if the user stops exercising for more than 1 minute (target duration) during exercise, the electronic device will activate screaming mode and emit a high-decibel warning sound until the user resumes exercise, and then it will re-enter the normal exercise monitoring state.

[0145] It is understandable that the volume, frequency and duration of the target warning sound can be set within a certain range according to the needs of the target user to achieve the best supervision effect.

[0146] For example, in a home fitness scenario, this function can provide supervision and guidance like a personal trainer for users who exercise at home. Users do not need to worry about interrupting exercise due to lack of self-discipline. The "scream if you don't exercise" mode can promptly urge them to resume exercise. The limb recognition guidance function can help users correct their movements and avoid sports injuries, making home fitness more scientific and efficient. During rehabilitation training, patients need to exercise strictly in accordance with the rehabilitation plan. The "scream if you don't exercise" mode of the electronic device can supervise patients to continue rehabilitation exercises. Its limb recognition technology can determine whether the patient's movements meet the rehabilitation requirements and provide timely voice guidance, which will help patients better complete rehabilitation training and accelerate physical recovery.

[0147] Optionally, after detecting that the target user's current exercise program is completed, the screaming-without-exercise mode may be automatically turned off.

[0148] and / or

[0149] S511: When the current exercise state is fatigue, a target encouragement message is sent.

[0150] Please refer to the following Figure 6, which is a flow chart of another motion recognition method provided by an exemplary embodiment of the present application. Figure 6 As shown, the motion recognition method may include but is not limited to the following steps:

[0151] S601: Receive a designated exercise type input by a target user.

[0152] Specifically, before the target user starts exercising, the electronic device can receive the designated exercise type corresponding to the current exercise input by the target user. After the exercise recognition function on the electronic device is turned on, the target user needs to perform the designated exercise type within the shooting range of its camera. The exercise types that the target user can specify include, but are not limited to, dance, whole-body exercise (such as, but not limited to, rope skipping, jumping jacks, high-leg running, etc.), strength training (such as, but not limited to, push-ups, burpees, plank support, etc.), etc.

[0153] S602: Acquire a target human motion image corresponding to a target user.

[0154] Specifically, the above S602 is consistent with the above S201 and will not be repeated here.

[0155] S603 , performing limb recognition on the target human motion image using a specified limb recognition algorithm or a specified limb recognition model corresponding to the specified motion type, to obtain target limb recognition data corresponding to the target user.

[0156] Specifically, the above-mentioned designated motion types are different, and the corresponding designated limb recognition algorithms or designated limb recognition models are different, or the limb motion parameters captured during the corresponding limb recognition process are different.

[0157] For example, when the designated exercise type is whole-body exercise, since the limb movements of whole-body exercises such as skipping rope, jumping jacks, and high-leg running are large and obvious, the corresponding designated limb recognition algorithm or designated limb recognition model can be used to clearly capture the motion trajectory and movement changes of various parts of the target human body, thereby accurately identifying the sports items and evaluating the accuracy of the movements.

[0158] When the designated exercise type is strength training, while the range of motion for strength training exercises like push-ups, burpees, and planks is relatively fixed, limb posture and joint angle changes are crucial. Therefore, the corresponding designated limb recognition algorithm or model can be used to accurately identify and analyze these key areas, effectively determining whether the user's movements are standard and providing accurate guidance.

[0159] When the designated exercise type is dance, some simple dance moves, such as basic aerobic dance moves, have a certain rhythmic and regular limb movement, and usually involve coordinated movements of the whole body. Therefore, the corresponding designated limb recognition algorithm or designated limb recognition model can be used to specifically capture the rhythm of the movement and the coordination of the limbs, accurately identify dance movements, and evaluate the user's movement smoothness and coordination. For example, when performing simple dance moves such as left and right arm swings and turns, the user's movements can be judged to determine whether they conform to dance standards by identifying parameters such as the limb's movement direction, angle, and speed, providing users with auxiliary support for dance learning and practice.

[0160] S604: Perform a motion evaluation on the target user based on the target physiological parameter data and the target limb recognition data during the target user's motion process to obtain a target motion evaluation result.

[0161] Specifically, the above S604 is consistent with the above S203 and will not be repeated here.

[0162] Please refer to the following Figure 7 , which is a structural diagram of a motion recognition device provided by an embodiment of the present application. Figure 7 As shown, the motion recognition device 700 includes:

[0163] An acquisition module 710 is configured to acquire a target human motion image corresponding to a target user;

[0164] The limb recognition module 720 is used to perform limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user;

[0165] The motion evaluation module 730 is used to perform motion evaluation on the target user based on the target physiological parameter data and target limb recognition data during the target user's motion process to obtain a target motion evaluation result; the target motion evaluation result includes the current motion state of the target user.

[0166] In one possible implementation, the limb recognition module 720 is specifically configured to:

[0167] Based on the target motion environment data corresponding to the target user, limb recognition is performed on the target human motion image to obtain target limb recognition data corresponding to the target user.

[0168] In one possible implementation, the target limb recognition data includes a sequence of first limb recognition results within a current time period; the sequence of first limb recognition results is composed of first limb recognition results corresponding to multiple frames of target human motion images acquired within the current time period, arranged in chronological order of acquisition; and the target physiological parameter data includes a sequence of first physiological parameters of the target user's current exercise within the current time period.

[0169] The above-mentioned motion evaluation module 730 is specifically used to: determine the current motion state of the above-mentioned target user based on the above-mentioned first limb recognition result sequence and the above-mentioned first physiological parameter sequence, and determine the current motion intensity information of the above-mentioned target user based on the above-mentioned first limb recognition result sequence; match the above-mentioned first limb recognition result sequence with a preset limb motion data sequence corresponding to at least one preset motion action, and determine the current motion action of the above-mentioned target user in the above-mentioned current time period based on the corresponding limb motion matching result; compare the current motion intensity information corresponding to the above-mentioned current motion action with the preset motion intensity information, and determine the current motion standard evaluation result corresponding to the above-mentioned current motion action based on the corresponding motion intensity comparison result; determine the current motion project of the above-mentioned target user based on the target motion action sequence during the current motion process of the above-mentioned target user; determine the current motion quality evaluation result corresponding to the above-mentioned current motion project based on the target motion intensity information and preset motion intensity information corresponding to each target motion action in the above-mentioned current motion project.

[0170] In a possible implementation, the motion assessment module 730 is specifically configured to:

[0171] The current physiological parameter change information of the target user is determined based on the above-mentioned first physiological parameter sequence, the preset physiological parameter threshold corresponding to the target user and / or the second physiological parameter sequence within the initial time period of the target user's current exercise; the current limb movement change information of the target user is determined based on the above-mentioned first limb recognition result sequence, the preset limb movement data sequence corresponding to the current movement of the target user and / or the second limb recognition result sequence within the above-mentioned initial time period; the current exercise state of the target user is determined based on the above-mentioned current physiological parameter change information and / or the above-mentioned current limb movement change information.

[0172] In a possible implementation, the motion recognition device 700 further includes:

[0173] An environmental data acquisition module is used to acquire target motion environment data corresponding to the target user using an environmental sensor;

[0174] The exercise intensity adjustment module is used to adjust the preset exercise intensity information corresponding to the current exercise action based on the target difference information between the target exercise environment data and the preset exercise environment data range when the target exercise environment data exceeds the preset exercise environment data range.

[0175] In a possible implementation, the target motion evaluation result further includes at least one of the following: a current action standard evaluation result and a current motion quality evaluation result corresponding to the target user;

[0176] The motion recognition device 700 further includes:

[0177] An action guidance module, configured to generate and issue current action guidance information corresponding to the target user based on the current action standard evaluation result; and / or

[0178] An exercise quality reminder module, configured to generate and send a current exercise quality reminder message corresponding to the target user based on the current exercise quality assessment result; and / or

[0179] an alert module configured to activate a screaming mode to emit a target alert sound if the current exercise program of the target user is not completed and the current exercise state is a stopped exercise state and the current duration corresponding to the stopped exercise state exceeds a target duration, and to deactivate the screaming mode to stop emitting the target alert sound after the current exercise state changes from a stopped exercise state to an exercise state; and the target emission volume corresponding to the target alert sound is greater than or equal to a target value; and / or

[0180] The exercise encouragement module is used to send a goal encouragement message when the current exercise state is a fatigue state.

[0181] In a possible implementation, the motion recognition device 700 further includes:

[0182] a receiving module configured to receive a target warning tone configuration operation input by the target user; the target warning tone configuration operation carries target configuration parameters selected by the target user for the target warning tone; the target configuration parameters include at least one of the following: a target type corresponding to the target warning tone, a target emission volume, a target emission duration, a target trigger frequency, and a target trigger condition;

[0183] The warning sound configuration module is used to respond to the target warning sound configuration operation and configure the target warning sound corresponding to the target user's exercise process based on the target configuration parameters.

[0184] In a possible implementation, the motion recognition device 700 further includes:

[0185] A voice command acquisition module is used to acquire the target voice command issued by the target user;

[0186] The mode activation module is used to respond to the above-mentioned target voice command to activate the screaming mode when not moving; the above-mentioned screaming mode when not moving is used to activate the above-mentioned screaming mode when the above-mentioned target user is in a state of stopping movement and the current duration corresponding to the stop movement state exceeds the above-mentioned target duration.

[0187] In a possible implementation, the motion recognition device 700 further includes:

[0188] An exercise type receiving module, configured to receive the specified exercise type input by the target user;

[0189] The limb recognition module 720 is specifically used for:

[0190] The target human motion image is subjected to limb recognition using the designated limb recognition algorithm or designated limb recognition model corresponding to the designated motion type to obtain target limb recognition data corresponding to the target user. Different designated motion types correspond to different limb motion parameters captured during the limb recognition process.

[0191] The division of the modules in the above-described motion recognition device is for illustrative purposes only. In other embodiments, the motion recognition device can be divided into different modules as needed to perform all or part of the functions of the above-described motion recognition device. The various modules in the motion recognition device provided in the embodiments of this specification can be implemented in the form of a computer program. This computer program can be run on a terminal, server, or image acquisition device. The program modules comprising this computer program can be stored in a memory of the terminal, server, or image acquisition device. When executed by a processor, this computer program implements all or part of the steps of the motion recognition method described in the embodiments of this specification.

[0192] See next Figure 8 , which is a structural diagram of an electronic device provided by an exemplary embodiment of this specification. Figure 8 As shown, the electronic device 800 may include: at least one processor 810 , at least one communication bus 820 , a user interface 830 , at least one network interface 840 , and a memory 850 .

[0193] The communication bus 820 may be used to implement connection and communication among the above components.

[0194] The user interface 830 may include a display screen (Display) and a camera (Camera), and the optional user interface 830 may also include a standard wired interface and a wireless interface.

[0195] The network interface 840 may optionally include a Bluetooth module, a Near Field Communication (NFC) module, a Wireless Fidelity (Wi-Fi) module, and the like.

[0196] The processor 810 may include one or more processing cores. The processor 810 utilizes various interfaces and circuits to connect various components within the electronic device 800. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 850, and accessing data stored in the memory 850, the processor 810 performs various functions and processes data for the routing electronic device 800. Optionally, the processor 810 may be implemented using at least one hardware form factor selected from the group consisting of a digital signal processing (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). The processor 810 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU primarily processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing the content displayed on the display; and the modem handles wireless communications. It is understood that the modem may not be integrated into the processor 810 and may be implemented as a separate chip.

[0197] Among them, the memory 850 may include a random access memory (RAM) or a read-only memory (ROM). Optionally, the memory 850 includes a non-transitory computer-readable medium. The memory 850 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 850 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 receiving function, a motion recognition function, a limb recognition 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 850 may also be optionally at least one storage device located away from the aforementioned processor 810. As Figure 8 As shown, the memory 850 as a computer storage medium may include an operating system, a network communication module, a user interface module, and program instructions.

[0198] In some possible embodiments, the electronic device 800 may be the aforementioned Figure 7 In the motion recognition device 700 shown, the processor 810 can be used to call program instructions stored in the memory 850 and perform the following operations:

[0199] Obtain a target human body motion image corresponding to the target user; perform limb recognition on the above target human body motion image to obtain target limb recognition data corresponding to the above target user; perform motion evaluation on the above target user based on the target physiological parameter data and target limb recognition data during the above target user's motion process to obtain a target motion evaluation result; the above target motion evaluation result includes the current motion state of the above target user.

[0200] In some possible embodiments, when the processor 810 performs the limb recognition on the target human motion image and obtains the target limb recognition data corresponding to the target user, it is specifically configured to perform:

[0201] Based on the target motion environment data corresponding to the target user, limb recognition is performed on the target human motion image to obtain target limb recognition data corresponding to the target user.

[0202] In some possible embodiments, the target limb recognition data includes a first limb recognition result sequence within the current time period; the first limb recognition result sequence is composed of the first limb recognition results corresponding to each of the multiple frames of target human motion images acquired within the current time period, arranged in chronological order of acquisition time; the target physiological parameter data includes the first physiological parameter sequence of the target user's current motion within the current time period; the processor 810 executes the target physiological parameter data and target limb recognition data based on the target user's motion process, performs motion evaluation on the target user, and when obtaining the target motion evaluation result, is specifically used to execute: determining the current motion state of the target user based on the first limb recognition result sequence and the first physiological parameter sequence, and determining the current motion state of the target user based on the first limb recognition result sequence, and determining the current motion state of the target user based on the first limb recognition result sequence, and determining the current motion state of the target user based on the first physiological parameter ... The method comprises the following steps: determining the current motion intensity information of the target user by sequence; matching the first limb recognition result sequence with a preset limb motion data sequence corresponding to at least one preset motion action, and determining the current motion action of the target user in the current time period based on the corresponding limb motion matching result; comparing the current motion intensity information corresponding to the current motion action with the preset motion intensity information, and determining the current motion standard evaluation result corresponding to the current motion action based on the corresponding motion intensity comparison result; determining the current motion project of the target user based on the target motion action sequence during this motion of the target user; and determining the current motion quality evaluation result corresponding to the current motion project based on the target motion intensity information and the preset motion intensity information corresponding to each target motion action in the current motion project.

[0203] In some possible embodiments, when the processor 810 determines the current motion state of the target user based on the first limb recognition result sequence and the first physiological parameter sequence, it is specifically configured to perform:

[0204] The current physiological parameter change information of the target user is determined based on the above-mentioned first physiological parameter sequence, the preset physiological parameter threshold corresponding to the target user and / or the second physiological parameter sequence within the initial time period of the target user's current exercise; the current limb movement change information of the target user is determined based on the above-mentioned first limb recognition result sequence, the preset limb movement data sequence corresponding to the current movement of the target user and / or the second limb recognition result sequence within the above-mentioned initial time period; the current exercise state of the target user is determined based on the above-mentioned current physiological parameter change information and / or the above-mentioned current limb movement change information.

[0205] In some possible embodiments, the processor 810 performs the above-mentioned motion evaluation of the target user based on the target physiological parameter data and the target limb recognition data during the motion process of the target user, and before obtaining the target motion evaluation result, is also used to perform: obtaining the target motion environment data corresponding to the target user using the environmental sensor; when the target motion environment data exceeds the preset motion environment data range, adjusting the preset motion intensity information corresponding to the current motion action based on the target difference information between the target motion environment data and the preset motion environment data range.

[0206] In some possible embodiments, the target motion evaluation result further includes at least one of the following: a current action standard evaluation result and a current motion quality evaluation result corresponding to the target user;

[0207] The processor 810 executes the target physiological parameter data and target limb recognition data of the target user during exercise, performs exercise evaluation on the target user, and after obtaining the target exercise evaluation result, is further used to execute: generating and issuing current action guidance information corresponding to the target user based on the current action standard evaluation result; and / or, generating and issuing current exercise quality reminder information corresponding to the target user based on the current exercise quality evaluation result; and / or, when the current exercise project of the target user is not completed, if the current exercise state is a stop exercise state and the current duration corresponding to the stop exercise state exceeds the target duration, the screaming mode is activated to issue a target warning sound, and after the current exercise state changes from a stop exercise state to an exercise state, the screaming mode is turned off to stop issuing the target warning sound; the target emission volume corresponding to the target warning sound is greater than or equal to the target value; and / or, when the current exercise state is a fatigue state, a target encouragement message is issued.

[0208] In some possible embodiments, the processor 810 is further configured to execute:

[0209] Receive a target warning sound configuration operation input by the target user; the target warning sound configuration operation carries the target configuration parameters selected by the target user for the target warning sound; the target configuration parameters include at least one of the following: the target type corresponding to the target warning sound, the target emission volume, the target emission duration, the target trigger frequency, and the target trigger condition; in response to the target warning sound configuration operation, configure the target warning sound corresponding to the target user during exercise based on the target configuration parameters.

[0210] In some possible embodiments, before executing the above-mentioned activation of the screaming mode to emit the target warning sound, the processor 810 is further configured to execute:

[0211] Obtain the target voice command issued by the above-mentioned target user; in response to the above-mentioned target voice command, turn on the screaming mode when not moving; the above-mentioned screaming mode when not moving is used to start the above-mentioned screaming mode when the above-mentioned target user is in a stopped state and the current duration corresponding to the stopped state exceeds the above-mentioned target duration.

[0212] In some possible embodiments, before performing the limb recognition on the target human motion image to obtain the target limb recognition data corresponding to the target user, the processor 810 is further configured to perform:

[0213] Receive the specified motion type of the target user input.

[0214] When the processor 810 performs the limb recognition on the target human motion image and obtains the target limb recognition data corresponding to the target user, it is specifically used to perform: using the specified limb recognition algorithm or specified limb recognition model corresponding to the specified motion type to perform limb recognition on the target human motion image and obtain the target limb recognition data corresponding to the target user; wherein, the limb motion parameters captured during the corresponding limb recognition process are different for different specified motion types.

[0215] The present application also provides a computer storage medium storing instructions that, when executed on a computer or processor, cause the computer or processor to perform one or more steps of any of the aforementioned methods. If the various components of the motion recognition device are implemented as software functional units and sold or used as independent products, they may be stored in the aforementioned storage medium.

[0216] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program product. The above-mentioned computer program product includes one or more computer instructions. When the above-mentioned computer program instructions are loaded and executed on a computer, the above-mentioned process or function according to the embodiment of the present application is generated in whole or in part. The above-mentioned computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The above-mentioned computer instructions can be stored in a computer-readable storage medium or transmitted via the above-mentioned computer-readable storage medium. The above-mentioned computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The above-mentioned computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The above-mentioned available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0217] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When executed, the program can include the processes of the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks. The technical features of this embodiment and the implementation scheme can be combined in any manner unless they conflict.

[0218] The above embodiments are merely descriptions of preferred embodiments of the present application and do not limit the scope of the present application. Without departing from the design spirit of the present application, various modifications and improvements made to the technical solutions of the present application by ordinary technicians in this field should fall within the scope of protection determined by the claims of the present application.

Claims

1. A motion recognition method, characterized in that: The method comprises: Obtaining a target human motion image corresponding to a target user; Performing limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user; Based on the target physiological parameter data and target limb recognition data of the target user during exercise, a motion evaluation is performed on the target user to obtain a target motion evaluation result; the target motion evaluation result includes the current motion state of the target user.

2. The method according to claim 1, characterized in that The performing limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user includes: Based on the target motion environment data corresponding to the target user, limb recognition is performed on the target human motion image to obtain target limb recognition data corresponding to the target user.

3. The method according to claim 1, characterized in that The target limb recognition data includes a sequence of first limb recognition results within a current time period; the first limb recognition result sequence is composed of first limb recognition results corresponding to multiple frames of target human motion images acquired within the current time period, arranged in chronological order of acquisition time; the target physiological parameter data includes a sequence of first physiological parameters of the target user within the current time period of the current exercise; The step of performing a motion evaluation on the target user based on the target physiological parameter data and the target limb recognition data during the target user's motion process to obtain a target motion evaluation result includes: Determine the current motion state of the target user based on the first limb recognition result sequence and the first physiological parameter sequence, and determine the current motion intensity information of the target user based on the first limb recognition result sequence; matching the first limb recognition result sequence with a preset limb motion data sequence corresponding to at least one preset motion action, and determining a current motion action of the target user within the current time period based on the corresponding limb motion matching results; Comparing the current exercise intensity information corresponding to the current exercise action with the preset exercise intensity information, and determining the current action standard evaluation result corresponding to the current exercise action based on the corresponding exercise intensity comparison result; Determining the current sport of the target user based on the target sport action sequence of the target user during the current sport; The current exercise quality evaluation result corresponding to the current exercise project is determined based on the target exercise intensity information corresponding to each target exercise action in the current exercise project and the preset exercise intensity information.

4. The method according to claim 3, characterized in that The determining the current motion state of the target user based on the first limb recognition result sequence and the first physiological parameter sequence includes: Determining current physiological parameter change information of the target user based on the first physiological parameter sequence, the preset physiological parameter threshold corresponding to the target user, and / or the second physiological parameter sequence within the initial time period of the target user's current exercise; Determining current limb movement change information of the target user based on the first limb recognition result sequence, the preset limb movement data sequence of the target user corresponding to the current movement action, and / or the second limb recognition result sequence within the initial time period; The current motion state of the target user is determined based on the current physiological parameter change information and / or the current limb movement change information.

5. The method according to claim 3, characterized in that Before performing a motion evaluation on the target user based on the target physiological parameter data and the target limb recognition data during the target user's motion process and obtaining a target motion evaluation result, the method further includes: Acquiring target motion environment data corresponding to the target user using an environmental sensor; In a case where the target motion environment data exceeds a preset motion environment data range, the preset motion intensity information corresponding to the current motion action is adjusted based on target difference information between the target motion environment data and the preset motion environment data range.

6. The method according to claim 1, characterized in that The target motion evaluation result further includes at least one of the following: a current action standard evaluation result and a current motion quality evaluation result corresponding to the target user; After performing a motion evaluation on the target user based on the target physiological parameter data and the target limb recognition data during the target user's motion process and obtaining a target motion evaluation result, the method further includes: Generate and issue current action guidance information corresponding to the target user based on the current action standard evaluation result; and / or Generate and send current exercise quality reminder information corresponding to the target user based on the current exercise quality evaluation result; and / or In the case where the current exercise program of the target user is not completed, if the current exercise state is a stopped exercise state and the current duration corresponding to the stopped exercise state exceeds the target duration, the screaming mode is activated to emit a target warning sound, and after the current exercise state changes from the stopped exercise state to the exercising state, the screaming mode is turned off to stop emitting the target warning sound; the target emission volume corresponding to the target warning sound is greater than or equal to the target value; and / or When the current exercise state is a fatigue state, a target encouragement message is sent.

7. The method according to claim 6, characterized in that The method further comprises: Receive a target warning sound configuration operation input by the target user; the target warning sound configuration operation carries target configuration parameters selected by the target user for the target warning sound; the target configuration parameters include at least one of the following: a target type corresponding to the target warning sound, a target emission volume, a target emission duration, a target trigger frequency, and a target trigger condition; In response to the target warning sound configuration operation, a target warning sound corresponding to the target user's exercise process is configured based on the target configuration parameters.

8. A motion recognition device, characterized in that: The device comprises: An acquisition module, used to acquire a target human motion image corresponding to a target user; A limb recognition module is used to perform limb recognition on the target human motion image to obtain target limb recognition data corresponding to the target user; The motion evaluation module is used to perform motion evaluation on the target user based on the target physiological parameter data and target limb recognition data during the target user's motion process to obtain a target motion evaluation result; the target motion evaluation result includes the current motion state of the target user.

9. An electronic device, characterized in that: include: processor and memory; wherein, The processor is connected to the memory; the memory is used to store executable program code; The processor reads the executable program code stored in the memory to run a program corresponding to the executable program code, so as to execute the method steps according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that The computer storage medium stores a plurality of instructions, which are suitable for being loaded by a processor and executing the method steps according to any one of claims 1 to 7.

Images