KR102423421B1 - Method of mornitoring home training using three dimensional modeling and server performing the same - Google Patents

Abstract

A home training monitoring server through three-dimensional modeling according to an embodiment of the present invention includes a three-dimensional image data receiving unit for receiving three-dimensional image data generated based on image data among data about an athlete performing home training, the image An audio data receiving unit for receiving audio data from among the image data captured by a photographing device, the three-dimensional image data and the audio data are analyzed to divide the three-dimensional image data and the audio data for each predetermined movement posture, and then the divided An exercise posture tag allocator for allocating different exercise posture tags to each of the three-dimensional image data and audio data, an exerciser rating determiner for determining the exerciser class using the exerciser information received from the exerciser terminal, the exerciser class and exercise posture pattern By analyzing the exercise posture database in which the standard exercise posture pattern is stored correspondingly and the three-dimensional image data to which the exercise posture tag is assigned, the exercise posture pattern is extracted, and the exercise posture database corresponds to the exerciser grade and the exercise posture pattern and an exercise posture feedback information generator configured to read a standard exercise posture pattern of

Description

Home training monitoring method through three-dimensional modeling and a server running it

The present invention relates to a method for monitoring home training through three-dimensional modeling and a server for executing the same. According to an embodiment of the present invention, three-dimensional image data can be generated by applying a three-dimensional pose estimation method to image data photographed by a user performing home training, and through this, exercise of an athlete while protecting personal information of the athlete It relates to a home training monitoring method through three-dimensional modeling that can provide feedback on posture, and to a server executing the same.

Exercise is known to be an essential element to maintain health, and accordingly, many fitness centers, pilates academies, and yoga centers are being created to promote the health of modern people.

However, it may be difficult for busy modern people to consistently go to a fitness center where exercise time is set to exercise. For this reason, interest in home training, which is an exercise at home, is increasing, and exercise contents or home training applications on YouTube are increasing.

However, in the case of existing video content or applications for home training, in many cases, the appearance of a trainer or an avatar teaching an exercise motion is provided from one angle.

Therefore, it may be difficult for the exerciser to follow the correct motion when imitating the motion, and it is difficult to determine whether the exerciser is doing it properly when imitating the motion.

In addition, the disadvantage of home training is that because it is an exercise performed by an athlete alone, interest may easily disappear, and it may be difficult to sustain for a long time for people who do not have a strong will.
Korean Patent Application Laid-Open No. 10-2021-0120148 relates to a “exercise posture analysis system,” and a method for analyzing an exercise posture is disclosed.” However, a method for solving the above problems is not disclosed.

An object of the present invention is to provide a home training monitoring method through three-dimensional modeling for generating three-dimensional image data by applying a three-dimensional pose estimation method to image data for an athlete performing home training, and a server for executing the same .

In addition, an object of the present invention is to provide a home training monitoring method through 3D modeling that can provide feedback on the exercise posture of the exerciser while protecting the exerciser's personal information through 3D image data and a server executing the same do it with

In addition, an object of the present invention is to provide a home training monitoring method through three-dimensional modeling that enables analysis and observation of an exercise posture of an exerciser without temporal and spatial constraints, and a server executing the same.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention not mentioned may be understood by the following description, and will be more clearly understood by the examples of the present invention. It will also be readily apparent that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the appended claims.

The home training monitoring server through three-dimensional modeling for achieving this purpose includes a three-dimensional image data receiving unit that receives three-dimensional image data generated based on image data among data about an athlete performing home training, the image capturing device A voice data receiver for receiving voice data among the image data captured by An exercise posture tag allocator for allocating different exercise posture tags to each of the video data and audio data, an exerciser rating determiner for determining the exerciser class using the exerciser information received from the exerciser terminal, and a standard for each exerciser class and exercise posture pattern An exercise posture database in which the movement posture pattern is stored correspondingly and the three-dimensional image data to which the exercise posture tag is assigned are analyzed to extract an exercise posture pattern, and the exerciser grade and the standard corresponding to the exercise posture pattern from the exercise posture database and an exercise posture feedback information generating unit that reads the exercise posture pattern and generates exercise posture feedback information by comparing the exercise posture pattern with the standard exercise posture pattern.

In addition, the home training monitoring method through 3D modeling executed in the home training monitoring server through 3D modeling for achieving this purpose is a 3D image generated based on image data among data about an exerciser who executes home training. Receiving data, receiving audio data among the image data photographed by the image photographing device, analyzing the three-dimensional image data and the audio data, and dividing the three-dimensional image data and the audio data for each predetermined movement posture allocating different exercise posture tags to each of the divided three-dimensional image data and audio data, analyzing the three-dimensional image data to which the exercise posture tag is assigned and extracting an exercise posture pattern; Reading a standard exercise posture pattern corresponding to an athlete's grade and the exercise posture pattern from an exercise posture database in which a standard exercise posture pattern is stored corresponding to each exercise posture pattern, and comparing the exercise posture pattern and the standard exercise posture pattern and generating the exercise posture feedback information and providing it to the exerciser terminal.

According to the present invention as described above, the advantage of being able to observe the movement posture of the exerciser using the 3D image data generated by applying the 3D pose estimation method to the image data of the exerciser performing home training is have.

In addition, according to the present invention, there is an advantage in that it is possible to provide feedback on the exercise posture of the exerciser while protecting the exerciser's personal information through the 3D image data.

In addition, according to the present invention, there is an advantage that it is possible to analyze and observe the movement posture of the exerciser without temporal and spatial restrictions.

1 is a network configuration diagram illustrating a system for providing home training through three-dimensional modeling according to an embodiment of the present invention.
2 is a block diagram illustrating an internal structure of a home training monitoring server according to an embodiment of the present invention.
3 is a flowchart illustrating an embodiment of a home training monitoring method according to an embodiment of the present invention.
4 is an exemplary diagram for explaining a home training monitoring process according to an embodiment of the present invention.

The above-described objects, features and advantages will be described below in detail with reference to the accompanying drawings, and accordingly, those skilled in the art to which the present invention pertains will be able to easily implement the technical idea of the present invention. In describing the present invention, if it is determined that a detailed description of a known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description will be omitted. Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings. In the drawings, the same reference numerals are used to refer to the same or similar components.

1 is a network configuration diagram illustrating a system for providing home training through three-dimensional modeling according to an embodiment of the present invention.

Referring to FIG. 1 , a system for providing home training through three-dimensional modeling includes an image capturing device 100 , a three-dimensional modeling device 200 , a home training monitoring server 300 , a trainer terminal 400 , and an exerciser terminal 500 . includes

The image photographing device 100 generates data by photographing a surrounding environment including an athlete performing home training. Thereafter, the image capturing device 100 provides image data among the data to the 3D modeling apparatus 200 , and provides audio data to the home training monitoring server 300 .

The image photographing device 100 may be, for example, a monocular camera or a stereo camera. Accordingly, an image captured by the image capturing device 100 may include a plurality of frames.

The 3D modeling apparatus 200 is an apparatus for generating 3D image data through a 3D pose estimation method using the image data received from the image capturing device 100 .

First, the 3D modeling apparatus 200 extracts at least one region from image data captured by the image photographing device 100 .

For example, the 3D modeling apparatus 200 determines a portion (eg, an edge) in which a gradient change corresponding to a first differential result of pixel intensity of an image exceeds a preset criterion. ) may be extracted by searching for at least one region including . If necessary, the 3D modeling apparatus 200 may detect an edge by using a result of the first differentiation and/or the result of the second differentiation with respect to the pixel intensity of the image.

Then, the three-dimensional modeling apparatus 200 performs a pose of the image photographing apparatus 100 that is tracked based on two-dimensional feature points extracted from the image photographed by the image photographing apparatus 100 . (pose) Receive information. In this case, the pose information of the image capturing device 100 changes according to, for example, a key frame corresponding to the secondary image data and movement of the image capturing device 100 from the first position to the second position. It may include at least one of the amount of change in the pose.

Here, the “key frame” corresponds to two frames in which two-dimensional feature points are extracted from among general frames, and a certain number or more is matched between the two frames to a level at which the relative positions between images can be determined.

For example, it takes a lot of computation time to extract the position of the feature point, extract a descriptor, match the feature point using the descriptor, and calculate the pose of the camera. Accordingly, since it is difficult to perform the corresponding process for every image, in an embodiment, the above-described process may be performed using a key frame.

At this time, the two-dimensional feature points extracted from the photographed image data correspond to sparse feature points, and may be extracted by a feature-based VSLAM method for calculating geometric information about the camera position and the scene of the image. .

The feature-based VSLAM method may track 6-DOF movement information between frames corresponding to the image data based on the extracted feature points of the image data. The pose information of the camera may include, for example, 6 degree of freedom pose information.

Thereafter, the 3D modeling apparatus 200 estimates first depth information of the image data based on the pose information of the at least one region and the image capturing device 100 .

For example, the 3D modeling apparatus 200 may generate a key frame corresponding to an image according to the movement of the image capturing device 100 from a first position to a second position based on the pose information of the image capturing device 100 . You can decide whether to replace (replace) with a new key frame or to redefine (refine) the key frame. Based on the determination, the modeling apparatus may estimate either depth information of depth information on a new key frame or depth information on a redefined key frame as the first depth information of the image.

The 3D modeling apparatus 200 generates 3D point clouds for modeling a map corresponding to an image based on the first depth information. The map corresponding to the image may be expressed by a pose graph of key frames. The key frames may be composed of an image, an inverse depth map, and a variance of the inverse depth map. Depth uncertainty in monocular SLAM is not well modeled by Gaussian.

For this reason, a predetermined character and identifier (eg, an athlete identifier, etc.) are mapped and expressed using an inverse depth instead of the classic X, Y, and Z representation. The “inverse depth map” may correspond to a map in which features are mapped using inverse depth as described above.

In addition, the variance of the inverse depth map and the inverse depth map may be defined for a subset of pixels. The inverse depth map and the variance of the inverse depth map may correspond to all regions in an image in which a gradient change of pixel intensity exceeds a preset criterion.

The three-dimensional point cloud may include, for example, points having three-dimensional coordinates, that is, three-dimensional points, and the three-dimensional points in the point cloud may be irregularly configured.

The home training monitoring server 300 extracts an exercise posture pattern using the three-dimensional image data received from the three-dimensional modeling device 200, and analyzes the exercise posture pattern based on the pre-generated exercise posture pattern big data to exercise It is a server that generates posture feedback information.

To this end, the home training monitoring server 300 provides each of three-dimensional image data and audio data to the trainer terminal 400 , and receives exercise posture feedback information by the trainer terminal 400 and provides it to the exerciser terminal 500 . do.

First, the home training monitoring server 300 determines the athlete's grade based on the experience information for each exercise type from the athlete terminal 500 and the grade upgrade information determined by the exercise posture feedback information generator 360 .

In one embodiment, the home training monitoring server 300 determines the athlete's grade based on the experience information for each exercise type from the athlete terminal 500 when there is no predetermined athlete grade for the athlete. That is, the home training monitoring server 300 may determine the athlete's grade according to the age of experience for each type of exercise.

In another embodiment, the home training monitoring server 300 raises the athlete's grade based on the grade upgrade information generated according to the athlete's home training result when there is a predetermined athlete grade for the athlete.

In addition, the home training monitoring server 300 extracts some image data and some audio data from among the 3D image data and audio data, respectively, and provides them to the trainer terminal 400 . The reason for this is that the capacity of each of the 3D image data and the audio data is very large, so it is difficult for the trainer to check all of the large amount of data.

Therefore, the home training monitoring server 300 divides the three-dimensional image data and audio data for each exercise posture, and then provides the divided three-dimensional image data and audio data to the trainer terminal 400, and later the trainer terminal 400 When the exercise posture feedback information received from the data is big data, 3D image data and audio data are divided for each exercise posture, and then analyzed and processed inside the trainer terminal 400 without being provided.

First, the home training monitoring server 300 divides the three-dimensional image data and audio data for each exercise posture, and then allocates an exercise posture tag to each of the divided three-dimensional image data and the audio data and provides it to the trainer terminal 400 . .

In this case, the exercise posture tag may be determined as the name of a motion that can be operated in the corresponding exercise according to the type of exercise. For example, when the type of exercise is yoga, the exercise posture tag may be determined as an action name that can be operated in yoga.

In one embodiment, the home training monitoring server 300 analyzes the voice data and when a predetermined word instructing the posture change is extracted, the voice data and the 3D image data are divided until the predetermined word is extracted.

Then, the home training monitoring server 300 extracts a word indicating a predetermined exercise posture from the divided voice data, and then assigns an exercise posture tag to each of the voice data and the 3D image data and provides it to the trainer terminal 400 . do.

For example, the home training monitoring server 300 analyzes the voice data, and when a word instructing a posture change such as “next posture” or “this posture” is extracted from the voice data, the voice data and 3D image data are pre-recorded. It is divided up to the point at which the determined word is extracted. Then, after extracting the word “bow posture” indicating a predetermined exercise posture from the divided voice data, a “bow posture tag” is assigned to each of the audio data and the 3D image data.

After that, the home training monitoring server 300 extracts the exercise posture pattern from the three-dimensional image data to which the exercise posture tag is assigned, and compares the exercise posture pattern and the predetermined exercise posture pattern to match the exercise posture feedback information create

In one embodiment, the home training monitoring server 300 reads the athlete grade and the standard exercise posture pattern corresponding to the exercise posture pattern from the exercise posture database, compares the exercise posture pattern and the standard exercise posture pattern, and provides exercise posture feedback information create

In the above embodiment, the home training monitoring server 300 calculates the matching rate of the exercise posture pattern and the standard exercise posture pattern, and if the match rate is greater than or equal to a specific rate, the class upgrade information for upgrading the exerciser class to the next class is generated. By doing so, the athlete's grade is raised.

That is, the home training monitoring server 300 compares the exercise posture pattern and the standard exercise posture pattern to calculate the difference angle between the body parts used to execute the exercise posture, and then calculates the matching ratio according to the difference angle. . For example, in the case of yoga exercise posture “asana”, the difference angle of the heel can be calculated by comparing the exercise posture pattern and the standard exercise posture pattern, and then the agreement ratio according to the difference angle can be calculated.

The trainer terminal 400 analyzes the 3D image data and audio data received from the home training monitoring server 300 , and generates exercise posture feedback information as a result of the analysis and provides it to the home training monitoring server 300 . it is a terminal

The exerciser terminal 500 is a terminal that is owned by an exerciser executing home training and receives exercise posture feedback information from the home training monitoring server 300 .

The exerciser terminal 500 provides the home training monitoring server 300 with experience information for each exercise type for determining the exerciser grade, and exercise posture feedback for home training according to the exerciser class determined by the home training monitoring server 300 . information can be received.

2 is a block diagram illustrating an internal structure of a home training monitoring server according to an embodiment of the present invention.

Referring to FIG. 2 , the home training monitoring server 300 includes a three-dimensional image data receiver 310 , an audio data receiver 320 , an exercise posture tag allocator 330 , an athlete rating determiner 340 , and an exercise posture database. 350 and an exercise posture feedback information generating unit 360 .

The 3D image data receiving unit 310 receives 3D image data generated based on image data among data about an athlete who performs home training.

In this case, the 3D image data is an image generated based on the image data of an athlete performing home training captured by the image capturing device 100 . Since the 3D image data is an image in which an athlete on the image data is modeled through a character, it is possible to observe the movement posture of the athlete while protecting personal information of the athlete.

The audio data receiver 320 receives audio data among the image data captured by the image photographing device 100 . For example, the voice data may include the voice of an instructor provided by the home training providing device and the voice of an athlete who performs home training according to an image provided by the home training providing device.

The exercise posture tag allocator 330 analyzes the three-dimensional image data and the audio data, divides the three-dimensional image data and the audio data for each movement posture, and then allocates the exercise posture tag to each of the divided three-dimensional image data and the audio data. do.

In this case, the exercise posture tag may be determined as the name of a motion that can be operated in the corresponding exercise according to the type of exercise. For example, when the type of exercise is yoga, the exercise posture tag may be determined as an action name that can be operated in yoga.

In one embodiment, the exercise posture tag allocator 330 analyzes the voice data and when a predetermined word instructing the posture change is extracted, the voice data and the 3D image data are divided until the predetermined word is extracted. .

Then, the exercise posture tag allocator 330 extracts a word indicating a predetermined exercise posture from the divided voice data, and then allocates the exercise posture tag to each of the voice data and the 3D image data.

For example, the exercise posture tag allocator 330 analyzes the voice data and when words instructing posture change such as “next posture” and “this posture” are extracted from the voice data, the voice data and 3D image data are recalled. Splits up to the point at which a predetermined word is extracted. Then, after extracting the word “bow posture” indicating a predetermined exercise posture from the divided voice data, a “bow posture tag” is assigned to each of the audio data and the 3D image data.

The exerciser rating determining unit 340 determines the exerciser rating based on the experience information for each exercise type from the exerciser terminal 500 and the rating upgrade information determined by the exercise posture feedback information generating unit 360 .

In one embodiment, the athlete rating determiner 340 determines the athlete rating based on the experience information for each exercise type from the athlete terminal 500 when there is no predetermined athlete rating for the athlete. That is, the home training monitoring server 300 may determine the athlete's grade according to the age of experience for each type of exercise.

In another embodiment, when there is a predetermined athlete grade for the athlete, the athlete grade determination unit 340 raises the athlete grade based on the grade upgrade information generated according to the athlete's home training result.

In the exercise posture database 350 , standard exercise posture patterns are stored in correspondence with each exerciser class and exercise posture pattern. As described above, the reason that the standard exercise posture pattern is stored for each exerciser class and exercise posture pattern is that the accuracy of the exercise posture that the exerciser can take is different according to the exerciser class.

For example, in the case of yoga's exercise posture "asana", the standard exercise posture pattern requires the heel to be on the floor, but depending on the athlete's grade, the heel may not be on the floor. A standard exercise posture pattern with different falling angles will be stored.

The exercise posture feedback information generating unit 360 extracts the exercise posture pattern from the three-dimensional image data to which the exercise posture tag is assigned, compares the exercise posture pattern with a predetermined exercise posture pattern, and generates exercise posture feedback information according to whether they match. create

In an embodiment, the exercise posture feedback information generating unit 360 reads the athlete grade and the standard exercise posture pattern corresponding to the exercise posture pattern from the exercise posture database, and compares the exercise posture pattern and the standard exercise posture pattern to the exercise posture Generate feedback information.

In the above embodiment, the exercise posture feedback information generating unit 360 calculates a matching rate between the exercise posture pattern and the standard exercise posture pattern, and when the match rate is greater than or equal to a specific rate, grade upgrade information for upgrading the exerciser rating to the next level is provided to the athlete rating determining unit 340 .

That is, the exercise posture feedback information generating unit 360 compares the exercise posture pattern and the standard exercise posture pattern to calculate the difference angle between the body parts used to execute the exercise posture, and then calculates the matching ratio according to the difference angle. can

For example, in the case of yoga exercise posture “asana”, the difference angle of the heel can be calculated by comparing the exercise posture pattern and the standard exercise posture pattern, and then the agreement ratio according to the difference angle can be calculated.

3 is a flowchart illustrating an embodiment of a home training monitoring method according to an embodiment of the present invention.

Referring to FIG. 3 , the home training monitoring server 300 receives 3D image data generated based on the image data among data about an exerciser executing home training (step S310 ).

The home training monitoring server 300 receives audio data among the image data captured by the image capturing device (step S320).

The home training monitoring server 300 analyzes the three-dimensional image data and the audio data, and divides the three-dimensional image data and the audio data for each predetermined exercise posture (step S330).

The home training monitoring server 300 allocates different exercise posture tags to each of the divided 3D image data and audio data (step S340).

The home training monitoring server 300 extracts an exercise posture pattern by analyzing the three-dimensional image data to which the exercise posture tag is assigned (step S350).

The home training monitoring server 300 reads the standard exercise posture pattern corresponding to the exerciser grade and the exercise posture pattern from the exercise posture database in which the standard exercise posture pattern is stored corresponding to each exerciser class and exercise posture pattern (step S350) .

The home training monitoring server 300 compares the exercise posture pattern and the standard exercise posture pattern to generate exercise posture feedback information and provides it to the exerciser terminal (step S360).

4 is an exemplary diagram for explaining a home training monitoring process according to an embodiment of the present invention.

Referring to FIG. 4 , the image capturing device 100 generates data by photographing a surrounding environment including an athlete who performs home training. Thereafter, the image capturing device 100 provides image data among the data to the 3D modeling apparatus 200 , and provides audio data to the home training monitoring server 300 .

The three-dimensional modeling apparatus 200 generates three-dimensional image data based on image data among data about an athlete performing home training, and provides the three-dimensional image data to the home training monitoring server 300 .

The home training monitoring server 300 extracts some image data and some audio data from among 3D image data and audio data, respectively, and provides them to the trainer terminal 400 . The reason for this is that the capacity of each of the 3D image data and the audio data is very large, so it is difficult for the trainer to check all of the large amount of data.

Therefore, the home training monitoring server 300 divides the three-dimensional image data and audio data for each exercise posture, and then provides the divided three-dimensional image data and audio data to the trainer terminal 400, and later the trainer terminal 400 When the exercise posture feedback information received from the data is big data, 3D image data and audio data are divided for each exercise posture, and then analyzed and processed inside the trainer terminal 400 without being provided.

Although it has been described with reference to the limited embodiments and drawings, the present invention is not limited to the above embodiments, and various modifications and variations are possible from these descriptions by those of ordinary skill in the art to which the present invention pertains. Accordingly, the spirit of the present invention should be understood only by the claims described below, and all equivalents or equivalent modifications thereof will fall within the scope of the spirit of the present invention.

100: video recording device
200: 3D modeling device
300: home training monitoring server
310: 3D image data receiving unit
320: voice data receiver
330: exercise posture tag allocation unit
340: athlete class determination unit
350: exercise posture database
360: exercise posture feedback information generating unit
400: trainer terminal
500: exercise terminal

Claims (2)

a three-dimensional image data receiver for receiving three-dimensional image data generated based on image data among data about an athlete performing home training;
an audio data receiver configured to receive audio data from among the video data captured by the video photographing device;
After analyzing the 3D image data and the audio data to divide 3D image data and audio data for each predetermined exercise posture, a movement of assigning different exercise posture tags to each of the divided 3D image data and audio data Posture tag allocator;
an athlete rating determining unit for determining the athlete's grade by using the athlete information received from the athlete terminal;
an exercise posture database in which standard exercise posture patterns are stored correspondingly for each exercise class and exercise posture pattern; and
Extracting an exercise posture pattern by analyzing the three-dimensional image data to which the exercise posture tag is assigned, reading a standard exercise posture pattern corresponding to the exerciser grade and the exercise posture pattern from the exercise posture database, and the exercise posture pattern and standard Comprising an exercise posture feedback information generating unit that compares the exercise posture pattern to generate exercise posture feedback information,
The exercise posture tag allocation unit
When a predetermined word instructing posture change is extracted by analyzing the voice data, the voice data and 3D image data are divided up to the point in time at which the predetermined word is extracted, and a word indicating a predetermined movement posture from the divided voice data After extracting, a movement posture tag is assigned to each of the audio data and the 3D image data,
The athlete grade determination unit
When a predetermined athlete grade for the athlete does not exist, the athlete grade is determined based on the experience information for each exercise type from the athlete terminal, and when a predetermined athlete grade for the athlete exists, the grade generated according to the athlete's home training result Upgrade the athlete's grade based on the upgrade information,
The exercise posture feedback information generating unit
The exercise posture database reads the athlete class and the standard exercise posture pattern corresponding to the exercise posture pattern, compares the exercise posture pattern and the standard exercise posture pattern to generate exercise posture feedback information, and the exercise posture pattern and the standard exercise posture pattern After calculating the angle of difference between the body parts used to perform the exercise posture by comparing Characterized in providing information to the athlete rating determining unit
Home training monitoring server through 3D modeling.
In the home training monitoring method through 3D modeling executed in the home training monitoring server through 3D modeling,
Receiving the 3D image data generated based on the image data among the data for the exerciser performing home training;
receiving audio data from among the image data captured by the image photographing device;
dividing the 3D image data and the audio data for each predetermined exercise posture by analyzing the 3D image data and the audio data;
allocating different exercise posture tags to each of the divided 3D image data and audio data;
extracting an exercise posture pattern by analyzing the three-dimensional image data to which the exercise posture tag is assigned;
reading a standard exercise posture pattern corresponding to the athlete class and the exercise posture pattern from an exercise posture database in which standard exercise posture patterns are stored corresponding to each exerciser class and exercise posture pattern; and
Comparing the exercise posture pattern and the standard exercise posture pattern, generating exercise posture feedback information and providing it to an exerciser terminal,
Allocating different exercise posture tags to each of the divided 3D image data and audio data includes:
When a predetermined athlete grade for the athlete does not exist, the athlete grade is determined based on the experience information for each exercise type from the athlete terminal, and when a predetermined athlete grade for the athlete exists, the grade generated according to the athlete's home training result Including the step of upgrading the athlete's grade based on the upgrade information,
Comparing the exercise posture pattern and the standard exercise posture pattern, generating exercise posture feedback information and providing it to the exerciser terminal
reading a standard exercise posture pattern corresponding to an athlete class and the exercise posture pattern from the exercise posture database;
generating exercise posture feedback information by comparing the exercise posture pattern and the standard exercise posture pattern; and
After comparing the exercise posture pattern and the standard exercise posture pattern to calculate the difference angle between the body parts used to execute the exercise posture, calculate a coincidence ratio according to the difference angle, and if the coincidence rate is greater than or equal to a certain proportion, the athlete rating It characterized in that it comprises the step of providing the grade upgrade information for upgrading to the next grade so that the athlete's grade is upgraded
Home training monitoring method through 3D modeling.

Images