KR20240026830A - Training supprot apparatus interworking with weight training apparatus - Google Patents

Abstract

The present invention relates to an exercise support device that works in conjunction with a strength training device.
The exercise support device according to an embodiment of the present invention may extract an exercise object exercising in a currently connected strength training device from a captured image captured by a camera module by analyzing skeleton data of candidate objects.
Through this, when multiple candidate objects are detected in the exercise image, the user exercising on the currently linked strength training device can be accurately identified using the exercise type and skeleton data of the candidate objects.

Description

Exercise support device linked to strength training equipment {TRAINING SUPPROT APPARATUS INTERWORKING WITH WEIGHT TRAINING APPARATUS}

The present invention relates to an exercise support device that links with a muscle strength exercise device, and more specifically, to an exercise support device that displays a user's exercise image in linkage with a strength exercise device.

Generally, strength training equipment used for strength training is designed to repeat muscle relaxation and contraction by pushing or pulling a lever connected to a weight having a certain unit of weight.

Among these exercise devices, there are a variety of exercise devices, such as arm curl machines for bicep exercises, chest press or butterfly machines for chest exercises such as the pectoralis major, and pull up machines for muscle exercises. Exercise equipment exists.

Recently, as part of self-management, home training, which involves exercising to manage your health at home, your own querencia, rather than jogging or visiting a fitness center, has become an issue. In line with this issue, not only are various exercise devices for home training being proposed, but a new word, home training group, has even been coined.

As an example of a strength exercise device for home training, a strength exercise device using a wire is used. These strength exercise devices have a drive motor installed inside them, transfer the exercise load provided from the drive motor to the user through a wire, and perform strength exercise by pulling or releasing the wire.

Meanwhile, in strength training, you must exercise with the correct posture to increase the efficiency of the exercise, and if you continue to exercise with the wrong posture, it may have a negative effect on your body.

Accordingly, a method has been proposed that uses a camera module to capture a user exercising using a strength training device and displays the video on the screen. For example, an image of the user exercising is displayed on the screen of a mobile device such as a tablet or TV.

However, in the case of a fitness center, many users are exercising, making it difficult to identify themselves when they want to view only their own exercise videos on the screen. For example, if you position the camera module so that only you are displayed on the screen, you can view your own exercise video, but if the installation location in the fitness center is restricted, multiple users, including yourself, are displayed on the screen. You can’t just watch your own exercise videos.

In addition, when a large number of users are displayed on the screen, there is a problem in that the user's appearance is relatively small and it is difficult to accurately determine the user's exercise posture.

Even in the case of home training, users may want to see their own workouts displayed on the screen, even if other people in the house, such as their children, follow the exercises or perform other actions next to the strength training machine. .

The purpose of the present invention is to provide an exercise support device that works in conjunction with a strength training device and can identify a user exercising in the strength training device from a captured image.

Another object of the present invention is to provide an exercise support device that can accurately specify a user exercising on a currently linked strength training device when multiple exercise objects exist in an exercise image.

Another object of the present invention is to provide an exercise support device capable of more accurately specifying a user by applying one or more exercise object specification methods.

Another object of the present invention is to provide an exercise support device that can increase the concentration of exercise by allowing only users of the current strength exercise device to be displayed on the screen.

An exercise support device linked to a strength training device according to an embodiment of the present invention may extract an exercise object exercising in a currently connected strength training device from an image captured by a camera module by analyzing skeleton data of candidate objects.

In one embodiment, an exercise support device according to an embodiment of the present invention may be configured to include a candidate object extraction module. The candidate object extraction module may extract skeleton data of a plurality of candidate objects from an image captured by a camera module.

An exercise support device according to an embodiment of the present invention may be configured to include an exercise object extraction module. The exercise object extraction module may analyze the skeleton data of each candidate object and extract as the main exercise object one of the plurality of candidate objects corresponding to the type of exercise performed by the strength training device connected through the terminal communication unit.

An exercise support device according to an embodiment of the present invention may be configured to include a device management module. In one embodiment, the device management module may adjust the captured image so that the main motion object extracted by the motion object extraction module is displayed at a preset position on the screen and display it on the image display unit.

In one embodiment, the terminal communication unit may receive information about the type of exercise performed by the strength training device from the strength training device. Here, the motion object extraction module may extract the main motion object based on information about the exercise type received through the terminal communication unit.

As another example, the device management module may receive information about the type of exercise to be performed by the strength training device. Additionally, the exercise object extraction module may extract the main exercise object based on the exercise type input to the device management module.

The terminal communication unit according to an embodiment of the present invention may receive device exercise information from the muscle strength exercise device. Additionally, when a plurality of main motion objects corresponding to a motion type are extracted, the motion object extraction module may calculate skeleton motion information from the skeleton data of each main motion object. Additionally, the motion object extraction module may compare the skeleton motion information and the device motion information and extract one of the plurality of main motion objects as the final motion object.

In one embodiment, the motion object extraction module may calculate at least one of a device motion speed and a device motion cycle from device motion information. Additionally, the motion object extraction module may extract the final motion object by comparing at least one of the device motion speed and the device motion cycle with the skeleton motion information.

In one embodiment, the motion object extraction module may extract a motion skeleton corresponding to a motion type from skeleton data. Additionally, the motion object extraction module may calculate at least one of the skeleton motion speed and the skeleton motion cycle according to the movement of the motion skeleton. Additionally, the motion object extraction module may extract the final motion object by comparing it with the device motion speed or device motion cycle.

In one embodiment, in an embodiment of the present invention, the device management module may receive the device exercise speed according to the exercise type and transmit the device exercise speed to the strength exercise device through the terminal communication unit. Additionally, when a plurality of main motion objects corresponding to an exercise type are extracted, the motion object extraction module may extract a motion skeleton corresponding to the exercise type from the skeleton data of each main motion object.

Additionally, the motion object extraction module calculates the skeleton motion speed according to the movement of the motion skeleton, and compares the device motion speed and the skeleton motion speed to extract the final motion object.

In one embodiment, the device management module may enlarge and display the captured image so that it is displayed on the image display with the main motion object located in the central area of the screen.

In one embodiment, the exercise object extraction module may be configured to include an artificial intelligence model that is generated by learning a plurality of training skeleton data, each labeled with an exercise type, as learning data, and the exercise type as output data. Additionally, the artificial intelligence model can receive the skeleton data of each candidate object and determine the movement type of the candidate object.

The exercise support device linked to the strength training device according to the present invention has one or more of the following effects.

First, an exercise support device is provided that can accurately identify a user exercising on a strength training device currently connected to the exercise support device from a captured video.

Second, when multiple candidate objects are detected in an exercise image, an exercise support device is provided that can accurately identify a user exercising in a currently linked muscle exercise device using the exercise type and skeleton data of the candidate objects.

Third, when there are multiple candidate objects exercising with the same exercise type, the effect of accurately specifying the user is provided based on the device motion information of the strength training device and the skeleton exercise information extracted from the skeleton data.

Fourth, your exercise video is displayed in the central area of the screen of the video display unit, allowing you to exercise while accurately checking your exercise video, providing the effect of increasing your concentration on exercise.

1 is a diagram showing an example of operation of an exercise support device and a muscle strength exercise device according to an embodiment of the present invention.
Figure 2 is a control block diagram of a strength training device according to an embodiment of the present invention.
Figure 3 is a diagram showing an example of a first drive system and a second exercise system of a muscle strength exercise device according to an embodiment of the present invention.
Figure 4 is a control block diagram of an exercise support device according to an embodiment of the present invention.
Figures 5 and 6 are diagrams for explaining an example of skeleton data extraction of the candidate object extraction module of the exercise support device according to an embodiment of the present invention.
Figure 7 is a diagram showing an example of an exercise situation using a strength training device according to an embodiment of the present invention.
Figure 8 is a diagram showing an example of a final exercise object displayed on the image display unit of an exercise support device according to an embodiment of the present invention.
Figure 9 is a diagram showing another example of an exercise situation using a strength training device according to an embodiment of the present invention.
Figure 10 is a control flowchart of an exercise support device according to an embodiment of the present invention.

The advantages and features of the present invention and methods for achieving them will become clear by referring to the embodiments described in detail below along with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below and may be implemented in various different forms. The present embodiments are merely provided to ensure that the disclosure of the present invention is complete and to be understood by those skilled in the art. It is provided to fully inform those who have the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout the specification.

Figure 1 is a diagram showing an example of operation of the exercise support device 30 and the muscle strength exercise device 10 according to an embodiment of the present invention.

As an example, the muscle strength exercise device 10 according to an embodiment of the present invention has a structure for muscle strength exercise in the form of pulling or releasing a pair of wires 120.

When described with reference to FIG. 1 , the strength training device 10 may be configured to include a device main body 110. In an embodiment of the present invention, it is assumed that the device body 110 has a mat shape. With the device main body 110 seated on the floor indoors or outdoors, the user climbs on top of the device main body 110 and uses a pair of wires 120 extending externally from the device main body 110 to apply muscle strength. Take exercising as an example.

Here, a gripping mechanism for a user to grip and exercise may be connected to the pair of wires 120. Figure 1 shows an example in which a handle bar 130 with wires 120 connected to both ends is installed. In addition, grips such as D-grips can be connected to each of the wires 120 on both sides, and of course, various types of gripping devices for exercise can be connected.

Figure 2 is a control block diagram of the muscle strength exercise machine 10 according to an embodiment of the present invention.

When described with reference to FIG. 2, the muscle strength exercise machine 10 according to an embodiment of the present invention may be configured to include a load supply module 150.

The load supply module 150 according to an embodiment of the present invention may generate an exercise load and provide the generated exercise load to the user through a pair of wires 120.

In an embodiment of the present invention, it is assumed that the load supply module 150 includes a first drive motor 151 and a second drive motor 152. The first drive motor 151 and the second drive motor 152 may independently generate exercise load and provide the generated exercise load through each wire 120.

Through this, in a state where the first drive motor 151 and the second drive motor 152 output corresponding exercise loads, the pair of wires 120 move in the same direction, as in the squat exercise shown in FIG. Movement of being pulled and released becomes possible.

In addition, with a D-grip or the like connected to each wire 120, it is also possible to pull and release a pair of wires 120 in opposite directions. In addition, one-handed movement using only one wire 120 is also possible.

In the embodiment shown in FIG. 2, the first drive motor 151 and the second drive motor 152 independently generate exercise load, but a pair of wires is generated using one drive motor and a differential mechanism. Of course, it can be implemented to provide the same exercise load as (120).

Meanwhile, the muscle strength exercise device 10 according to an embodiment of the present invention may be configured to include a position detection unit 160.

In one embodiment, the position detection unit 160 may detect the extraction position of the wire 120. As described above, the first drive motor 151 and the second drive motor 152 are independently connected to each wire 120, and the position detection unit 160 is connected to the first position detection unit 161 and the second drive motor 152. As an example, it is configured to include a second position detection unit 162.

Here, the first position detection unit 161 detects the drawing position of one wire 120 of the pair of wires 120, and the second position detection unit 162 detects the drawing position of the other wire 120 of the pair of wires 120. The extraction position of the wire 120 is detected.

In one embodiment, the first position detection unit 161 and the second position detection unit 162 each directly or indirectly detect the rotation amount of the first drive motor 151 and the second drive motor 152, The extraction position of the wire 120 can be detected. For example, the first position detection unit 161 and the second position detection unit 162 may detect the rotation of the first drive motor 151 and the second drive motor 152, respectively.

As another example, the load supply module 150 according to an embodiment of the present invention includes a first drive system and a first drive system for transmitting the rotational force of the first drive motor 151 and the second drive motor 152 to each wire 120. 2 It can be configured to include a drive system. In one embodiment, each of the first drive system and the second drive system may be configured to include a bobbin, timing belts 151b and 152b, and transmission pulleys 151a and 152a, as shown in FIG. 3.

The transmission pulleys 151a and 152a are connected to the first drive motor 151 or the second drive motor 152 through timing belts 151b and 152b, and the first drive motor 151 or the second drive motor 152 ) can rotate in synchronization with the rotation. In addition, the bobbin is coaxially coupled to the transmission pulleys 151a and 152a with the wire 120 wound thereon, and rotates in synchronization with the rotation of the transmission pulleys 151a and 152a, thereby forming the timing belts 151b and the transmission pulley. The exercise load of the first drive motor 151 or the second drive motor 152 transmitted through (151a, 152a) is transmitted to each wire 120.

Here, the first position detection unit 161 and the second position detection unit 162 operate in the forward and reverse directions of the timing belts 151b and 152b according to the rotation of the first drive motor 151 and the second drive motor 152. By detecting the movement, the extraction position of each wire 120 can be detected.

The configuration of the first position detection unit 161 and the second position detection unit 162 according to the embodiment of the present invention is not limited to the above-described embodiment, and detects the position of the wire 120 that moves according to the user's movement. It can be composed of various forms.

Again, referring to FIG. 2 , the muscle strength exercise device 10 according to an embodiment of the present invention may be configured to include a device communication unit 140. In one embodiment, the device communication unit 140 may be connected to the exercise support device 30 through a communication network such as Bluetooth or Wi-Fi.

The muscle strength exercise device 10 according to an embodiment of the present invention may be configured to include a device control unit 170.

The device control unit 170 can control the entire function of the muscle strength exercise device 10 according to an embodiment of the present invention. In one embodiment, the device control unit 170 may control the first drive motor 151 and the second drive motor 152 so that the exercise load is output as a target load set by the user.

In an embodiment of the present invention, the device control unit 170 transmits device movement information based on the extraction position of each wire 120 detected by the first position detection unit 161 and the second position detection unit 162 to the device communication unit. It is transmitted to the exercise support device 30 through 140, which will be described later.

Meanwhile, the exercise support device 30 according to an embodiment of the present invention is connected to the strength exercise device 10 through a communication network and can interoperate with the strength exercise device 10.

In the embodiment shown in FIG. 1, it is assumed that the exercise support device 30 is provided in the form of a tablet. In addition, it can be implemented in various forms such as mobile devices such as smart phones, laptop computers, desktop computers, and TVs.

Figure 4 is a control block diagram of the exercise support device 30 according to an embodiment of the present invention.

When described with reference to FIG. 4 , the exercise support device 30 according to an embodiment of the present invention may be configured to include a terminal communication unit 340.

In one embodiment, the terminal communication unit 340 may be connected to the device communication unit 140 of the strength training device 10 through a communication network such as Bluetooth or Wi-Fi. Additionally, the terminal communication unit 340 may receive device exercise information from the muscle exercise device 10 .

The exercise support device 30 according to an embodiment of the present invention may be configured to include a camera module 320. In one embodiment, the camera module 320 is arranged to face the direction of the strength training device 10 and can capture a user exercising on the strength training device 10.

In an embodiment of the present invention, it is assumed that the camera module 320 includes a CMOS image sensor and is installed on the front of the exercise support device 30 to capture a user's exercise image. As another example, a separate camera module 320 for taking images is connected to the exercise support device 30, and the video captured from the camera module 320 is sent to the exercise support device 30 through wired or wireless communication. You can receive it.

The exercise support device 30 according to an embodiment of the present invention may be configured to include an image display unit 330. In one embodiment, when the exercise support device 30 according to an embodiment of the present invention is provided in the form of a tablet or a mobile device such as a smart phone, an image display unit 330 may be formed on one front side. As another example, if the exercise support device 30 according to an embodiment of the present invention is a laptop computer or a desktop computer, the image display unit 330 may be provided in the form of a monitor, and if it is a TV, the TV screen itself may be the image display unit 330. It can be.

Here, the image display unit 330 may be implemented as a liquid crystal display (LCD) panel, a light emitting diode (LED) panel, an organic light emitting diode (OLED) panel, etc., which will be described later. The graphical user interface provided by the device management module 313 may be displayed on the screen.

The exercise support device 30 according to an embodiment of the present invention may be configured to include a candidate object extraction module 311. Here, the candidate object extraction module 311 may extract skeleton data of a plurality of candidate objects from the captured image captured by the camera module 320.

In one embodiment, the candidate object extraction module 311 may determine a moving object in a captured image and extract the object as a candidate object. Additionally, the candidate object extraction module 311 may extract a plurality of key points preset for the candidate object as skeleton data.

Figure 5 is a diagram showing an example of the candidate object extraction module 311 according to an embodiment of the present invention extracting skeleton data of a user performing a squat exercise, and Figure 6 is a diagram illustrating a candidate object extraction module according to an embodiment of the present invention ( 311) is a diagram showing an example of extracting skeleton data of a user performing cable arm curl exercise.

As shown in Figures 5 and 6, the positions of the head, neck, hips, and multiple joints of a candidate object, a person, can be extracted as skeleton data. In one embodiment, the plurality of joint locations may include both shoulders, both elbows, both hands or wrists, both hips, both knees, and both ankles. In one embodiment, the candidate object extraction module 311 may extract skeleton data of the candidate object using an algorithm such as the OpenVINO Library.

The exercise support device 30 according to an embodiment of the present invention may be configured to include an exercise object extraction module 312. The exercise object extraction module 312 analyzes the skeleton data of each candidate object and extracts the candidate object exercising in the strength training device 10 connected through the terminal communication unit 340 as the main exercise object among the candidate objects. .

In one embodiment, the exercise object extraction module 312 may extract as the main exercise object one corresponding to the type of exercise performed by the strength training device 10 connected through the terminal communication unit 340.

As shown in Figures 5 and 6, the skeleton data extracted by the candidate object extraction module 311 has different aspects depending on the type of exercise the user is performing on the strength training device 10. Accordingly, the exercise object extraction module 312 analyzes the skeleton data of the candidate objects and selects a Hugo object having skeleton data of the type of exercise being performed on the strength training device 10 currently connected through the terminal communication unit 340. It can be extracted as the main movement object.

In an embodiment of the present invention, it is assumed that the motion object extraction module 312 includes an artificial intelligence-based artificial intelligence model. As an example, an artificial intelligence model may be created based on deep learning-based algorithms such as Convolution Neural Network (CNN), K-Nearest neighborhood (KNN), and Deep Neural Network (DNN).

An artificial intelligence model can be created by learning a plurality of training skeleton data, each labeled with an exercise type, as learning data, and the exercise type as output data. Therefore, when the skeleton data of each candidate object is input, the artificial intelligence model can determine the movement type of each candidate object through an inference process.

In one embodiment, multiple sets of skeleton data for one exercise type may be applied as skeleton data for learning. As an example, time-series skeleton data for one exercise cycle can be learned as a set of skeleton data for learning.

In addition, the artificial intelligence model can receive skeleton data for each candidate object in time series from the candidate object extraction module 311 and infer the movement type of the candidate object.

Figure 7 is a diagram showing an example of an exercise situation using the strength training device 10 according to an embodiment of the present invention, taking as an example a situation in which two candidate objects are extracted from a captured image captured by the camera module 320. there is.

In Figure 7, candidate object A is performing a cable arm curl exercise, and candidate object B is performing a squat exercise. Here, as shown in FIGS. 5 and 6, skeleton data of each candidate object may be extracted by the candidate object extraction module 311.

As described above, the artificial intelligence model of the exercise object extraction module 312 receives the skeleton data of each candidate object, sets the exercise type of candidate object A as a cable arm curl exercise, and the exercise type of candidate object B as a squat exercise. comes to inference.

Here, when the user inputs the squat exercise as the exercise type through the strength exercise machine 10 or the exercise support device 30, the exercise object extraction module 312 extracts the candidate object B as the main exercise object. .

Again, referring to FIG. 4 , the exercise support device 30 according to an embodiment of the present invention may be configured to include a device management module 313.

In one embodiment, the device management module 313 is connected to the strength exercise device 10 through the terminal communication unit 340, and performs strength exercise using the strength exercise device 10 through linkage with the strength exercise device 10. It can provide various functions necessary for.

For example, the user may input the target load through the terminal input unit 360, and the device management module 313 may transmit the target load entered by the user to the strength exercise device 10 through the terminal communication unit 340. .

Additionally, when the user inputs an exercise type through the terminal input unit 360, the device management module 313 may transmit information about the exercise type to the strength training device 10 through the terminal communication unit 340. Additionally, the device management module 313 may transmit information about the input exercise type to the exercise object extraction module 312.

Here, the device management module 313 may display a graphic user interface to support muscle strength exercise through the strength exercise device 10 on the image display unit 330. Additionally, the device management module 313 may process input signals input by the user through a graphical user interface displayed on the image display unit 330, for example, a target load or exercise type selection signal.

The exercise support device 30 according to an embodiment of the present invention may be configured to include a terminal control unit 350.

In one embodiment, the terminal control unit 350 may include a hardware component and a software component that perform the overall functions of the exercise support device 30. For example, the terminal control unit 350 may include various operating systems (OS), middleware, platforms, and various applications, as well as program codes for signal processing and control, memory for storing them, and a processor for executing them. .

Here, the candidate object extraction module 311, the exercise object extraction module 312, and the device management module 313 according to an embodiment of the present invention are implemented in the form of a single management application 310, and the exercise support device 30 ) can be installed. And, when the management application 310 is executed, its functions can be executed through linkage with the terminal control unit 350.

The device management module 313 according to an embodiment of the present invention adjusts the captured image so that the main motion object extracted by the motion object extraction module 312 is displayed at a preset position on the screen and displays it on the video display unit 330. You can.

In one embodiment, as shown in FIG. 8, the size and position of the captured image can be adjusted so that it is located in the central area of the screen of the graphic user interface provided by the device management module 313.

As shown in FIG. 7, with two people captured in the captured image, the candidate object extraction module 311 extracts skeleton data of a plurality of candidate objects, and the motion object extraction module 312 extracts the plurality of candidate objects. When the main motion object is extracted from among the candidate objects, the device management module 313 zooms in and shifts the captured image so that the main motion object is located in the central area of the screen as shown in FIG. 8 and displays it on the image display unit 330. I do it.

Meanwhile, Figure 9 is a diagram showing another example of an exercise situation using the strength training device 10 according to an embodiment of the present invention, in which three candidate objects are extracted from the captured image captured by the camera module 320. is given as an example. Unlike the example shown in FIG. 7, this is an example of a situation in which three candidate objects A, B, and C are all performing a squat exercise.

Therefore, even if the skeleton data of each candidate object is extracted by the candidate object extraction module 311, the exercise type extracted by the exercise object extraction module 312 may all be squat exercises.

Accordingly, as an example, the terminal communication unit 340 of the exercise support device 30 according to an embodiment of the present invention receives device exercise information from the muscle strength exercise device 10.

As described above, the device control unit 170 of the strength exercise device 10 uses the information about the extraction position of the wire 120 detected by the position detection unit 160 as device exercise information to connect the device communication unit 140 to the device communication unit 140. It can be transmitted to the exercise support device 30 through.

Here, the motion object extraction module 312 can calculate skeleton motion information from the skeleton data of each main motion object when a plurality of main motion objects corresponding to the motion type are extracted, as shown in FIG. 9. there is.

Additionally, the motion object extraction module 312 may compare the skeleton motion information and the device motion information and extract one of the plurality of main motion objects as the final motion object.

In one embodiment, the motion object extraction module 312 may extract at least one of the device motion speed and the device motion period from the device motion information. Referring to FIG. 8 , even if main exercise objects A, B, and C all perform squat exercise, their mutual exercise speed or exercise cycle may be different.

Accordingly, the motion object extraction module 312 may calculate the motion speed of each main motion object based on the device motion information, that is, the extraction position of the wire 120. Additionally, the exercise object module can calculate an exercise cycle, which is one exercise cycle, based on device motion information.

In one embodiment, the motion object extraction module 312 may extract a motion skeleton corresponding to a motion type from skeleton data. Then, the motion object extraction module 312 calculates at least one of the skeleton motion speed and the skeleton motion cycle according to the movement of the motion skeleton, and compares the device motion speed or the device motion cycle to extract the final motion object.

For example, in the case of a squat exercise, when the positions of the wrist and elbow are extracted from the skeleton data as the exercise skeleton, they may correspond to the extraction position of the wire 120. Therefore, based on changes in the positions of the wrist and elbow, the skeleton movement speed or skeleton movement cycle can be calculated. As another example, in the case of the cable arm curl exercise shown in FIG. 7, if the position of the wrist is extracted from the skeleton data as the exercise skeleton, it can correspond to the extraction position of the wire 120.

In another embodiment, when the strength training device 10 according to an embodiment of the present invention is implemented in a form capable of running exercise, the device exercise speed may be input. For example, when the user selects treadmill exercise as the exercise type, the device exercise speed at which the user wants to run can be input through the terminal input unit 360.

Additionally, the device management module 313 may transmit the device exercise speed to the strength training device 10 through the terminal communication unit 340. The device control unit 170 may control the running speed of the strength training device 10 based on the device exercise speed received through the device communication unit 140.

In one embodiment, the exercise object extraction module 312 extracts a plurality of main exercise objects corresponding to an exercise type, for example, when a plurality of main exercise objects in progress on a treadmill exercise are extracted, each main exercise The movement skeleton corresponding to the movement type can be extracted from the object's skeleton data.

As an example, the exercise object extraction module 312 may extract the knees and ankles of the treadmill exercise as exercise skeletons and calculate the skeleton movement speed according to the movement of the corresponding exercise skeleton.

Additionally, the motion object extraction module 312 may compare the machine motion speed and the skeleton motion speed input through the terminal input unit 360 to extract one of the plurality of main motion objects as the final motion object.

Hereinafter, the control process of the exercise support device 30 according to an embodiment of the present invention will be described with reference to FIG. 10.

First, as described above, when the user selects his/her exercise type and starts exercising, an exercise image is captured by the camera module 320 (S80).

Additionally, the candidate object extraction module 311 may extract candidate objects from the captured image captured by the camera module 320 (S18) and extract skeleton data of the candidate objects.

Then, the motion object extraction module 312 analyzes the skeleton data of the candidate objects and extracts the main motion object from the candidate object corresponding to the previously registered motion type (S82).

Here, the motion object extraction module 312 determines whether a plurality of main motion objects corresponding to the exercise type are detected (S83), and if only one main motion object is extracted, extracts the main motion object as the final motion object. It is done (S84).

On the other hand, when a plurality of main motion objects are extracted in step S83, the motion object extraction module 312 extracts a motion skeleton from the skeleton data of each main motion object and calculates the skeleton motion speed based on the movement of the motion skeleton. It is done (S85).

In addition, the exercise object extraction module 312 calculates the device exercise speed based on the device motion information received from the strength exercise device 10 through the terminal communication unit 340, that is, the extraction position of the wire 120, and skeleton exercise By comparing the speed and the device movement speed, the main movement object with the skeleton movement speed corresponding to the device movement speed is extracted as the final movement object (S86).

When the final motion object is extracted as described above, the device management module 313 adjusts the captured image captured by the camera module 320 so that, as shown in FIG. 8, the final motion object is located in the central area of the screen. The captured image is displayed on the image display unit 330 (S88).

Although embodiments of the present invention have been described above with reference to the attached drawings, the present invention is not limited to the above embodiments and can be manufactured in various different forms, and can be manufactured in various different forms by those skilled in the art. It will be understood by those who understand that the present invention can be implemented in other specific forms without changing its technical spirit or essential features. Therefore, the embodiments described above should be understood in all respects as illustrative and not restrictive.

10: Strength exercise device 110: Device body
120: wire 130: handle bar
140: device communication unit 150: load supply module
151: first driving motor 152: second driving motor
151a, 152a: Transmission pulley 151b, 152b: Timing belt
160: position detection unit 161: first position detection unit
162: second position detection unit 170: device control unit
30: exercise support device 310: management application
311: Candidate object extraction module 312: Movement object extraction module
313: device management module 320: camera module
330: video display unit 340: terminal communication unit
350: terminal control unit 360: terminal input unit

Claims (9)

a terminal communication department;
a camera module;
an image display unit that displays captured images captured by the camera module;
a candidate object extraction module that extracts skeleton data of a plurality of candidate objects from the captured image;
An exercise object extraction module that analyzes the skeleton data of each candidate object and extracts one of the plurality of candidate objects corresponding to a type of exercise performed by a muscle strength exercise device connected through the terminal communication unit as a main exercise object; ;
An exercise support device linked to a strength training device, comprising a device management module that adjusts the captured image so that the main exercise object is displayed at a preset location on the screen and displays it on the image display unit. According to paragraph 1,
The terminal communication unit receives information about the type of exercise performed by the strength training device from the strength training device;
The exercise object extraction module extracts the main exercise object based on information about the exercise type received through the terminal communication unit. According to paragraph 1,
The device management module receives information about the type of exercise to be performed by the strength training device;
The exercise object extraction module extracts the main exercise object based on the exercise type input to the device management module. According to paragraph 1,
The terminal communication unit receives device exercise information from the strength exercise device;
The motion object extraction module is
When a plurality of main motion objects corresponding to the motion type are extracted, skeleton motion information is calculated from the skeleton data of each main motion object,
An exercise support device linked to a strength training device, characterized in that one of the plurality of main exercise objects is extracted as a final exercise object by comparing the skeleton exercise information with the device motion information. According to clause 4,
The motion object extraction module is
An exercise support device linked to a strength training device, wherein at least one of a device motion speed and a device motion cycle is calculated from the device motion information, and the final exercise object is extracted by comparison with the skeleton motion information. According to clause 5,
The motion object extraction module is
Extract a movement skeleton corresponding to the movement type from the skeleton data,
Calculate at least one of a skeleton movement speed and a skeleton movement cycle according to the movement of the exercise skeleton, and extract the final movement object by comparing it with the device movement speed or the device movement cycle. Exercise support devices. According to paragraph 1,
The device management module receives the device exercise speed according to the exercise type and transmits the device exercise speed to the strength exercise device through the terminal communication unit;
The motion object extraction module is
When a plurality of main motion objects corresponding to the motion type are extracted, a motion skeleton corresponding to the motion type is extracted from the skeleton data of each main motion object,
Calculate the skeleton movement speed according to the movement of the movement skeleton,
An exercise support device linked to a muscle strength exercise device, characterized in that the final exercise object is extracted by comparing the device movement speed with the skeleton movement speed. According to paragraph 1,
The device management module enlarges and displays the captured image so that it is displayed on the image display unit with the main exercise object located in the central area of the screen. According to paragraph 1,
The exercise object extraction module is generated by learning using a plurality of training skeleton data labeled with each exercise type as learning data and the exercise type as output data, and receives the skeleton data of each candidate object to extract the candidate object. An exercise support device linked to a strength training device, comprising an artificial intelligence model that determines the type of exercise.

Images