KR102372613B1 - Method, apparatus and computer-readable medium for correctting posture through overlay analysis of image - Google Patents

Abstract

The present invention relates to a method of correcting an exercise posture through an overlap analysis of images. According to an embodiment of the present invention, a first image input as a reference image for an exercise posture from a user terminal and a comparison image for the first image an overlapping image generating step of generating an overlapping image by overlapping the input second image; In the overlapping image, by comparing the degree of overlap with respect to the motion posture of the first object included in the first image and the second object included in the second image, the overlapping error of the second object with respect to the first object is analyzed. error analysis step; and a feedback information providing step of providing, to the user terminal, feedback information for minimizing the overlap error with respect to the motion postures of the first object and the second object based on the result of the overlap error analysis.

Description

Exercise posture correction method, apparatus and computer-readable recording medium through superimposition analysis of images

The present invention relates to a method of correcting an exercise posture through an image overlap analysis, and specifically, by superimposing a reference image and a comparison image for the exercise posture, errors in exercise postures such as golf, baseball, yoga and Pilates are analyzed and to provide feedback information according to error analysis to support the correction of the user's exercise posture.

When exercise is performed with the correct posture, exercise can increase exercise efficiency by stimulating the desired muscles to strengthen the body and prevent injury. It is very high and the use of the muscles to be trained is reduced, so fatigue is high, but exercise efficiency is extremely low, so it is very important to perform exercise with the right posture.

Therefore, people who start exercising hire a personal trainer to receive coaching on their exercise posture, but hiring a personal trainer is not only expensive, but also has to visit the exercise center at the appointed time. There was a big limit to the enemy constraint.

For this reason, in recent years, for people who exercise alone, posture correction techniques for various sports are being developed. device) as an example.

Specifically, in the prior art described above, a technique for correcting a golf swing posture is disclosed, and by providing a practice device having a plate for guiding the club, a foot seat part, and a user's upper body and lower body correction part, golf A technique for correcting a swing posture is disclosed.

However, the above-mentioned prior art has a problem that is cumbersome to use because it is required to install a complicated golf swing practice device whenever a user wants to practice a golf swing motion, and furthermore, the above-mentioned prior art is a backswing, a backswing top And since only posture correction limited to the motion for the down swing is possible, the range of application thereof is extremely limited in that the posture correction for the entire golf swing motion cannot be performed.

Accordingly, an object of the present invention is to superimpose a reference image for an exercise posture and a comparison image, analyze an error on the exercise posture, and provide feedback information according to the error analysis to support correction of a user's exercise posture There is this.

In order to achieve the above object, the golf swing posture correction method through the overlap analysis of images implemented in a computing device including one or more processors and a main memory for storing instructions executable in the processor according to an embodiment of the present invention is , an overlapping image generating step of generating an overlapping image by superimposing a first image input as a reference image for an exercise posture from a user terminal and a second image input as a comparison image with respect to the first image; In the overlapping image, by comparing the degree of overlap with respect to the motion posture of the first object included in the first image and the second object included in the second image, the overlapping error of the second object with respect to the first object is analyzed. error analysis step; and a feedback information providing step of providing, to the user terminal, feedback information for minimizing the overlap error with respect to the motion postures of the first object and the second object based on the result of the overlap error analysis.

In the above-described overlapping image generating step, an area including at least one of a body joint area and an exercise tool area of the first object included in the first image and the second object included in the second image is set as the overlap reference area. Thus, it is preferable to generate an overlapping image in which the first image and the second image are combined by correcting the size and direction.

In addition, before performing the above-described overlap error analysis step, from the superimposed image, the frame-by-frame motion posture for the first object and the second object for the image section corresponding to the motion motion is extracted, and the extracted motion posture for each frame It is desirable to apply to the deep learning algorithm so that one or more key postures for motor motions are detected.

In addition, the above-described overlapping error analysis step acquires XY coordinate parameters for one overlapping reference area of the first object and the second object, corresponding to the detected core posture change point, and moves with respect to the one overlapping reference area It is preferable to analyze the overlap error with respect to the motion posture by allowing the trajectory change value of the motion to be calculated and comparing the trajectory change values calculated for the first object and the second object.

In addition, in the above-described overlapping error analysis step, whenever an input to the second image is performed from the user terminal, the trajectory change of the second object is accumulated by calculating the trajectory change value of the second object included in the second image. It is desirable to derive the average value and variance value for each core posture of the user from the value, and to analyze the posture stability for each key posture of the user from the variance value.

In addition, the above-described feedback information providing step may include, through an external network, searching for an image about the exercise posture of a professional athlete having a similarity between the average value for each core posture of the user and a preset threshold or more, and recommending the searched image as the first image. It is desirable to provide

In addition, in the above-described overlapping error analysis step, with respect to the detected core posture, the switching speed between the core postures included in the exercise operation is calculated by using the time required to change the posture from the first core posture to the second core posture. It is preferable to analyze the overlap error for the exercise posture by comparing the switching speed between the core postures calculated for the first object and the second object.

In addition, the above-described feedback information providing step is one of the feedback information provided to the user terminal when an image of a professional athlete's exercise posture is input as the first image, and is applied to one or more exercise items extracted from the first image. It is desirable to provide purchase link information for

On the other hand, the golf swing posture correction apparatus through the overlap analysis of images implemented as a computing device including one or more processors and a main memory for storing instructions executable by the processor is input from a user terminal as a reference image for an exercise posture. an overlapping image generating unit generating an overlapping image by overlapping the first image and a second image input as a comparison image with respect to the first image; In the overlapping image, by comparing the degree of overlap with respect to the motion posture of the first object included in the first image and the second object included in the second image, the overlapping error of the second object with respect to the first object is analyzed. error analysis unit; and a feedback information providing unit that provides feedback information to the user terminal for minimizing the overlap error with respect to the motion postures of the first object and the second object based on the result of the overlap error analysis.

On the other hand, as a computer-readable recording medium, the computer-readable medium stores instructions for causing a computing device to perform the following steps: Input from a user terminal as a reference image for an exercise posture an overlapping image generating step of generating an overlapping image by overlapping the first image and a second image input as a comparison image with respect to the first image; In the overlapping image, by comparing the degree of overlap with respect to the motion posture of the first object included in the first image and the second object included in the second image, the overlapping error of the second object with respect to the first object is analyzed. error analysis step; and a feedback information providing step of providing, to the user terminal, feedback information for minimizing the overlap error with respect to the motion postures of the first object and the second object based on the result of the overlap error analysis.

According to an embodiment of the present invention, by superimposing the reference image and the comparison image for the exercise posture, the error on the exercise posture is analyzed in detail, and feedback information according to the error analysis is provided to correct the exercise posture of the user. There is a supportive effect.

In particular, taking golf as an example, the present invention enables the overall correction of the golf swing posture, that is, the golf swing posture from the address to the finish posture, to be performed, so that the correction effect on the user's golf swing posture can be increased. do.

In addition, according to an embodiment of the present invention, it is possible to provide results on the degree of improvement of the user's exercise behavior for a preset period using artificial intelligence, as well as provide information on the exercise posture of professional players and users. As it is possible to provide the analysis result for the difference, it is possible to provide personalized feedback to the user who wants to correct the exercise posture.

In addition, according to an embodiment of the present invention, since it is possible to provide information on the stability of each core posture of the user by calculating the average value and variance value for each core posture of the user related to the exercise motion, the user himself/herself without a separate coach It has the effect of being able to evaluate one's own proficiency.

In addition, according to an embodiment of the present invention, as a reference image used to correct a user's exercise posture, an image of a professional athlete's exercise movement similar to the current user's exercise posture is used, thereby efficiently improving the exercise posture. It has the effect of making corrections possible.

1 is a flowchart of a method for correcting an exercise posture through an image overlap analysis according to an embodiment of the present invention.
2 and 4 illustrate an example in which dynamic object recognition is performed in an image input from a user terminal according to an embodiment of the present invention.
3 and 5 are an example of a conceptual diagram for determining an overlap error by overlapping a first object of a first image and a second object of a second image according to an embodiment of the present invention;
6 is an example in which an error with respect to an exercise posture is analyzed according to a comparison of trajectory change values according to an embodiment of the present invention.
7 is an example in which posture stability for each core posture of a user is analyzed according to an embodiment of the present invention;
8 is an example in which feedback information on an exercise posture is provided to a user terminal according to an embodiment of the present invention.
9 is an example in which purchase information for an exercise item extracted from a first image is provided to a user terminal according to an embodiment of the present invention.
10 is a block diagram of an apparatus for correcting an exercise posture through an image overlap analysis according to an embodiment of the present invention.
11 is an example of an internal configuration of a computing device according to an embodiment of the present invention.

Hereinafter, various embodiments and/or aspects are disclosed with reference to the drawings. In the following description, for purposes of explanation, numerous specific details are set forth to provide a thorough understanding of one or more aspects. However, it will also be recognized by one of ordinary skill in the art that such aspect(s) may be practiced without these specific details. The following description and accompanying drawings set forth in detail certain illustrative aspects of one or more aspects. These aspects are illustrative, however, and some of various methods may be employed in the principles of the various aspects, and the descriptions set forth are intended to include all such aspects and their equivalents.

As used herein, “embodiment”, “example”, “aspect”, “exemplary”, etc. may not be construed as an advantage or advantage in any aspect or design described above over other aspects or designs. .

Also, the terms "comprises" and/or "comprising" mean that the feature and/or element is present, but excludes the presence or addition of one or more other features, elements, and/or groups thereof. should be understood as not

Also, terms including an ordinal number such as 1st, 2nd, etc. may be used to describe various components, but the components are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component. and/or includes a combination of a plurality of related listed items or any of a plurality of related listed items.

In addition, in the embodiments of the present invention, unless otherwise defined, all terms used herein, including technical or scientific terms, are those commonly understood by those of ordinary skill in the art to which the present invention belongs. have the same meaning. Terms such as those defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in an embodiment of the present invention, an ideal or excessively formal meaning is not interpreted as

The present invention relates to a method for correcting exercise posture through image overlap analysis, and by superimposing a reference image and a comparison image for the exercise posture, analyzing the error for the exercise posture, and providing feedback information according to the error analysis to provide feedback information according to the error analysis. The purpose is to support the correction of posture.

On the other hand, as a description of the present invention for achieving the above-described object, a more specific description of a method for correcting an exercise posture through image overlap analysis will be performed with reference to the accompanying drawings, one technical feature or invention A plurality of drawings will be referenced simultaneously to describe the components constituting the .

First, referring to FIG. 1 , there is shown a flowchart of a method for correcting an exercise posture through image overlap analysis according to an embodiment of the present invention.

1, in the present invention, a first image input as a reference image for an exercise posture from a user terminal and a second image input as a comparison image for the above-described first image are overlapped to generate an overlapped image. An overlapping image generation step ( S10 ) may be performed.

In this case, the above-described first image is a reference image for correcting the user's exercise posture, and among images captured directly from the user terminal, images downloaded through an external network, and images stored in a database linked to an application implementing a service It may be understood as a concept including at least one, and images according to sports may be used.

In addition, the exercise referred to in the present invention may be understood as a concept including a bare body exercise such as yoga, stretching, as well as an exercise using tools such as a golf club, a baseball bat, a ball, and a foam roller.

Meanwhile, the above-described second image is an image including a dynamic object for which exercise posture correction is to be performed, and may be understood as an image of a user's exercise posture.

At this time, in another embodiment of the present invention, the above-described first image recommendation may be performed to the user terminal, and in this case, the above-described first image recommendation is provided to the user terminal for information on the user's body information and an exercise item to be learned. It may be understood that this is accomplished by performing a request for input of information including at least one of information and preferred professional player information in advance.

That is, in the present invention, since an exercise image corresponding to the user's physical condition or desired to be learned by the user and an exercise image for the exercise posture of a professional player preferred by the user can be used as a reference image, the user It is possible to expect additional effects such as increase in service use by inducing learning motivation and giving interest in correcting exercise posture.

On the other hand, when the above-described step S10 is performed, the overlap of the first image and the second image is a body joint region and exercise tool of the first object included in the first image and the second object included in the second image. It may be understood that an overlapping image in which the first image and the second image are combined is generated by setting the area including at least one of the areas as the overlapping reference area to correct the size and direction between images or objects. .

Specifically, the above-described body joint region may be, for example, a region including at least any one of a neck, shoulder, elbow, wrist, pelvis, knee, and ankle, and the above-described exercise tool region may include golf In this case, the area related to the golf club will be an area including at least any one of the head, which is the part under the club that comes into contact with the ball when hitting the ball, the shaft which is the longest part of the club, and the handle, and when the sport is baseball, a baseball bat It may be understood that a region including at least one of a handle tip, a handle, and a striking surface is recognized as a related region.

Referring to FIG. 2 for a more detailed description, FIG. 2 shows an example in which dynamic object recognition is performed in an image input from a user terminal according to an embodiment of the present invention.

Specifically, the embodiment of FIG. 2 may be understood as an example in which object recognition corresponding to a swing motion of golf is performed among various sports, and in this case, 100 in FIG. It can be understood as a dynamic object extracted from an image, and 110 of FIG. 2 is to be understood as an example in which a body joint area of a dynamic object including at least one of a first object and a second object is detected and a golf swing posture is estimated. can

In this case, preferably, for a more detailed analysis of the golf swing posture, as shown in FIG. 2 , identification points of different colors are designated in regions corresponding to the body joint regions, and by connecting them to correspond to the body structure, the golf swing By extracting the posture, it would be desirable to improve the estimation accuracy for the golf swing posture of the dynamic object.

Meanwhile, in FIG. 4 , as an embodiment of another sports event of the present invention, an example in which object recognition corresponding to a pitching motion of baseball is performed is shown. Similarly, 200 in FIG. 4 denotes at least one of the first image and the second image. It can be understood as a dynamic object extracted from one image, and 210 of FIG. 4 is an example in which a body joint area of a dynamic object including at least one of a first object and a second object is detected and a baseball pitching motion posture is estimated can be understood

At this time, in the description of FIGS. 2 and 4 above, embodiments limited to golf and baseball among sports are described, but the present invention recognizes at least any one of a body joint area and an exercise tool area as an identification point, so that the posture of the object is Since the estimation is performed, it is possible to estimate the motion posture of the object in the image regardless of the exercise event, and accordingly, posture correction according to various exercise items can be performed, thereby expanding the range of application.

Returning to FIG. 1 and continuing the description, after the above-described step S10 is performed, in the generated superimposed image, a first object that is a dynamic object included in the first image and a dynamic object included in the second image An overlap error analysis step ( S20 ) of analyzing the overlap error of the second object with respect to the first object by comparing the degree of overlap with respect to the movement posture of the second object may be performed.

At this time, in a more preferred embodiment of the present invention, in the superimposed image generated in step S10 before performing the above-described step S20, the frame-by-frame movement posture for the first object and the second object for the image section corresponding to the movement motion A function that allows to be extracted may be further performed.

In detail, the above-described exercise posture corresponds to golf as an example, starting with the address posture in which the first posture is taken for the golf swing, the backswing posture in which the body rotation starts with the start of the club swing, and the apex posture of the backswing Backswing top position, the club down swing position towards the ball, the impact position where the ball and the club touch, the follow-through position where the arms are extended in the target direction by the centrifugal force generated by the swing after the impact position, and the end of the swing It is preferable to understand that up to the finish posture corresponding to

In addition, the present invention can apply the extracted frame-by-frame golf swing posture to a deep learning algorithm to detect one or more key postures for a golf swing motion. By comparing the errors of the golf swing postures of the two objects, correction of the golf swing posture can be performed.

3 and 5 for a more detailed description, in FIGS. 3 and 5, the overlapping error is analyzed by superimposing the first object of the first image and the second object of the second image in step S10 of FIG. 1 . An example of a concept that makes it possible is shown.

Specifically, 120 to 140 of FIG. 3 are an example for a case where the exercise event is golf, and 120 of FIG. 3 shows the golf swing posture of the first object extracted from the first image serving as the reference image. may be understood as , and reference numeral 130 of FIG. 3 may be understood as the golf swing posture of the second object extracted from the second image, which is a comparison image.

In addition, in 140 of FIG. 3 , as the first image and the second image overlap, the golf swing posture of the first object overlaps the golf swing posture of the second object, and thus an embodiment is shown in which the degree of overlap between the two objects is compared. can be known

As another embodiment, referring to FIG. 5 , the embodiments 220 to 240 of FIG. 5 are for a case in which the sports event is baseball, and 220 of FIG. 5 is extracted from the first image serving as the reference image. It may be understood as the baseball pitching posture of the first object, and 230 of FIG. 5 may be understood as the baseball pitching posture of the second object extracted from the second image, which is a comparison image.

In addition, 240 of FIG. 5 may be understood as an embodiment in which, as the first image and the second image overlap, the baseball pitching posture of the first object overlaps the baseball pitching posture of the second object, and thus the degree of overlap between the two objects is compared. will be.

That is, as previously discussed in the embodiments of FIGS. 3 and 5 , in the present invention, a technique for correcting the movement posture by comparing the overlapping error degree of the movement posture of the second object based on the movement posture of the first object is disclosed. In particular, in the present invention, through the overlap of the first image and the second image, it is possible to objectively determine how much the user's exercise posture is different from the posture of the reference image, thereby further reducing the overlapping error of the exercise posture. By being able to analyze closely, it is possible to expect the effect of remarkably increasing the correction precision of the exercise posture.

Referring back to FIG. 1 , in more detail with respect to the performance of the above-described step S20, various embodiments may exist in analyzing the overlap error in step S20 of the present invention.

As a first embodiment, in the above-described step S20, XY coordinate parameters for one overlapping reference area of the first object and the second object corresponding to the detected core posture change point are acquired, and the movement is performed on the one overlapping reference area. By allowing the trajectory change value of the motion to be calculated and comparing the trajectory change values calculated for the above-described first and second objects, the overlap error for the movement posture can be compared. Referring to FIG. 6 , in 300 of FIG. 6 , as an example of a golf swing motion, the trajectory change value calculated for the first object and the second object by setting the overlap reference area for the first object and the second object to the wrist An example of this derived is shown.

At this time, in the coordinate plane of 300 of FIG. 4 , 310 will be understood to indicate a trajectory change value using the XY coordinate parameter obtained from the first object, and 320 will be a trajectory change value using the XY coordinate parameter obtained from the second object. It will be understood as indicated.

That is, in the present invention, the overlap error of the motion posture of the second object with respect to the motion posture including the golf swing posture of the first object can be analyzed in detail by the above-described error analysis process.

Meanwhile, according to another embodiment of the present invention, whenever an input to the second image is performed from the user terminal, the trajectory change value of the second object included in the second image is calculated, and the second image accumulated on the database is calculated. 2 From the trajectory change value of the object, an average value and a variance value for each core posture of the user are derived, and an analysis of the posture stability for each core posture of the user may be performed from the derived variance value.

Specifically, in the present invention, the second image, which is a comparison image, is accumulated and managed in the database in the order of time input from the user terminal, and the user using the XY coordinate parameter of the second object obtained from the accumulated second images It is possible to derive the average value and variance value for each core posture of

That is, in the present invention, it is possible to grasp the average posture for each core posture of the user with respect to the movement from the above-mentioned average value, and it is possible to grasp the posture stability for each core posture of the user from the above-described dispersion value.

Meanwhile, in 400 of FIG. 7 , graphs of average values and variance values according to measured values of XY parameters obtained from the second object are shown.

At this time, in the graph of FIG. 7 , 410 will be understood to mean a parameter for the trajectory of the user's average exercise posture for one core posture, and the y-axis change value (Δy) will be understood to mean the degree of dispersion.

To explain in detail, as the y-axis change value is smaller, it is interpreted that the user's exercise posture does not change significantly, and it can be analyzed that the posture stability is high. It can be interpreted as low postural stability.

That is, in the present invention, by reducing the deviation of the user's exercise motion through the above-described analysis of posture stability, it is possible to perform training to make the user's exercise skill proficient.

Returning to FIG. 1 again, as a second embodiment of analyzing the overlap error in step S20 of the present invention, for the core posture detected in step S20, it is required to switch the posture from the first core posture to the second core posture From the time information, the switching speed between the core postures included in the exercise is calculated, and the overlapping error for the exercise posture can be compared by comparing the switching speed between the core postures calculated for the first object and the second object. there is.

That is, by performing the above-described second embodiment, even error analysis on the switching speed between core postures can be performed, so that more precise exercise posture correction can be performed.

In this case, in the performance of step S20, the aforementioned embodiments may be performed independently of each other, or one or more embodiments may be organically combined and performed simultaneously, but the present invention is not limited thereto.

On the other hand, after the above-described step S20 is performed, based on the result of the overlap error analysis in step S20, the above-described feedback information for minimizing the overlap error with respect to the motion posture of the first object and the second object is provided to the above-described user terminal. A step (S30) of providing feedback information to be provided may be performed.

At this time, when the above-described feedback information is golf, and a reference image and a comparison image for a golf swing motion are input, it is preferable that feedback information about the finish posture from the address posture corresponding to the golf swing motion is provided, When the sports event is baseball and a reference image and a comparison image for a baseball pitching motion are input, it may be desirable to provide feedback information on from a windup posture to a follow-through posture corresponding to the baseball pitching motion.

That is, it will be understood that the feedback information described above in the present invention provides feedback information for each exercise motion according to the exercise type, and is applied to various sports such as golf and baseball, as well as other sports, bowling, yoga, and Pilates as described above. It can be understood as applicable.

Meanwhile, examples of providing feedback information for a swing motion of golf and a pitching motion of baseball are illustrated in 500 and 510 of FIG. 8 , and a detailed description of the feedback information will be continued with reference to 500 and 510 .

First, referring to 500 of FIG. 8 , in the above-described feedback information, the golf swing postures of the first object and the second object are superimposed, so that the user can intuitively compare and view their golf swing postures 501 ) may be included.

At this time, the above-described posture information 501 may be provided as a still image or a dynamic image in which an overlapping screen of the first object and the second object captured as a still screen for each core posture is provided as a still image, and accordingly, the user can play his/her golf It is possible to receive and view an image that intuitively compares a swing posture with an object of a reference image (eg, a professional golfer), thereby increasing learning efficiency for posture correction.

In addition, it is natural that the above-described feedback information may further include correction tip information 502 for each core posture in addition to the posture information 501, and correction tip information for minimizing the overlap error between the first object and the second object will be understood to be provided.

On the other hand, in the feedback information on the baseball pitching motion shown in 510 of FIG. 8, similarly, the baseball pitching posture of the first object and the second object are overlapped to allow the user to intuitively compare and view their pitching posture ( 511) may be included, and correction tip information 512 for each key posture for a pitching motion in a process similar to a golf swing may also be included.

That is, in the present invention, feedback on a user's exercise posture according to an exercise item is provided in a user-customized manner, so that it is possible to derive a posture correction effect that is the same as hiring a personal trainer or improved compared to a personal trainer.

On the other hand, in another embodiment of the present invention, in performing the above-described step S30, an image of the exercise posture of a professional athlete having a similarity greater than or equal to a preset threshold to the average value for each core posture of the user derived earlier through an external network The search may be performed to provide feedback for recommending and providing the searched image as the first image.

That is, in performing this function, in taking an exercise posture for each exercise item, an image including an exercise posture of a professional player having an exercise posture similar to that of the user is recommended as a first image as a reference image, thereby determining the user's current exercise posture. It is possible to provide the effect of improving the exercise skill of the user in a short time without significantly correcting it.

Further, in another embodiment of the present invention, in the performance of step S30, when an image of a professional athlete's exercise posture is input as the first image serving as the reference image in step S10, among the feedback information provided to the user terminal As one, purchase link information for one or more exercise items extracted from the first image may be provided.

That is, as shown in the screen 600 of FIG. 9 , when a first image related to golf is input, at least one of a golf club, a caddy bag, a golf ball, and golf wear detected by object recognition in the above-described first image is selected. As it is possible to provide purchase link information for the included golf item to the user terminal side, by performing this function, shopping guide for the golf item can be provided to the user side, as well as natural advertisement for the golf item is performed, so that it is possible to provide the effect of providing a means of generating profits to the service provider side.

Overall, according to an embodiment of the present invention, according to an embodiment of the present invention, the reference image for the exercise posture and the comparison image are superimposed to closely analyze the error of the exercise posture, and feedback according to the error analysis There is an effect that can support the correction of the user's exercise posture by providing information.

In particular, for golf, for example, in the present invention, the overall posture of the golf swing operation, that is, the overall correction of the golf swing posture from the address to the finish posture can be performed, so that the correction effect on the user's golf swing posture can be increased. be able to

In addition, according to an embodiment of the present invention, it is possible to provide results on the degree of improvement of the user's exercise behavior for a preset period using artificial intelligence, as well as provide information on the exercise posture of professional players and users. As it is possible to provide the analysis result for the difference, it is possible to provide personalized feedback to the user who wants to correct the exercise posture.

In addition, according to an embodiment of the present invention, since it is possible to provide information on the stability of each core posture of the user by calculating the average value and variance value for each core posture of the user related to the exercise motion, the user himself/herself without a separate coach It has the effect of being able to evaluate one's own proficiency.

In addition, according to an embodiment of the present invention, as a reference image used to correct a user's exercise posture, an image of a professional athlete's exercise movement similar to the current user's exercise posture is used, thereby efficiently improving the exercise posture. It has the effect of making corrections possible.

As described above, although the embodiments have been described with reference to the limited embodiments and drawings, various modifications and variations are possible from the above description by those skilled in the art.

Meanwhile, FIG. 10 shows a configuration diagram of an exercise posture correction apparatus 10 through image overlap analysis according to an embodiment of the present invention, and in the following description, unnecessary implementation that overlaps with those of FIGS. 1 to 9 A description of the example will be omitted.

As shown in Fig. 10, the exercise posture correction apparatus 10 through image overlap analysis described in the present invention is preferably an overlapping image generating unit 11, an overlapping error analyzing unit 12, and a feedback information providing unit ( 13) may be included.

Specifically, the above-described superimposed image generating unit 11 superimposes a first image input from the user terminal 20 as a reference image for an exercise posture and a second image input as a comparison image for the above-described first image It performs a function to create a superimposed image.

In this case, the reference region for overlapping the above-described first image and the second image is at least one of a dynamic object included in the first image and the second image, that is, a body joint region and an exercise tool region of the first object and the second object. It will be understood that a reference region including

That is, it can be understood that the above-described superposed image generating unit 11 can perform all of the functions performed in step S10 of FIG. 1 , and in the present invention, by performing the functions of the above-described superposed image generating unit 11 , the reference Since the first image, which is an image, and the second image, which is a comparison image, are shown to be overlapped, there is an effect that an error with respect to an exercise posture can be more closely grasped.

Next, the above-described overlapping error analyzer 12 is configured to perform the analysis of a first object that is a dynamic object included in the above-described first image and a second object that is a dynamic object included in the above-described second image in the generated overlapping image. It performs a function of analyzing the overlap error of the above-described second object with respect to the above-described first object by comparing the degree of overlap with respect to the movement posture.

Specifically, the above-described overlap error analysis unit 12, the trajectory change value for each core posture calculated for the movement motions of the first object and the second object, the switching speed between the core postures, and the user's posture stability according to the passage of time It can be understood that the function to analyze the trend of change is performed.

As a result, it is preferable that the above-described overlap error analysis unit 12 performs all the functions performed in step S20 of FIG. 1 above, and the user's exercise posture is performed by the function of the overlap error analysis unit 12 described above. That is, taking the golf swing motion as an example, the overall correction of the golf swing posture from the address to the finish posture can be performed, so that the correction effect of the user's golf swing posture can be further increased.

Next, the above-described feedback information providing unit 13, based on the overlapping error analysis result derived from the above-described overlap error analysis unit 12, the overlap error for the motion posture of the first object and the second object described above. It is possible to perform a function of providing feedback information for minimizing the above to the user terminal 20 .

That is, it can be understood that the above-described feedback information providing unit 13 can perform all of the functions performed in step S30 of FIG. 1 above, and in the present invention, by performing the functions of the above-described feedback information providing unit 13, As it is possible to provide results on the degree of improvement of the user's exercise behavior for a preset period by using the There is an effect that it is possible to provide personalized feedback to a user who wants to correct an exercise posture.

As a result, according to an embodiment of the present invention, by superimposing the reference image and the comparison image for the exercise posture, the error on the exercise posture is analyzed in detail, and feedback information according to the error analysis is provided to provide the user's exercise posture It has the effect of supporting the correction of

In particular, taking golf as an example, the present invention enables the overall correction of the golf swing posture, that is, the golf swing posture from the address to the finish posture, to be performed, so that the correction effect on the user's golf swing posture can be increased. do.

In addition, according to an embodiment of the present invention, it is possible to provide results on the degree of improvement of the user's exercise behavior for a preset period using artificial intelligence, as well as provide information on the exercise posture of professional players and users. As it is possible to provide the analysis result for the difference, it is possible to provide personalized feedback to the user who wants to correct the exercise posture.

In addition, according to an embodiment of the present invention, since it is possible to provide information on the stability of each core posture of the user by calculating the average value and variance value for each core posture of the user related to the exercise motion, the user himself/herself without a separate coach It has the effect of being able to evaluate one's own proficiency.

In addition, according to an embodiment of the present invention, as a reference image used to correct a user's exercise posture, an image of a professional athlete's exercise movement similar to the current user's exercise posture is used, thereby efficiently improving the exercise posture. It has the effect of making corrections possible.

On the other hand, the user terminal 20 referred to in the present invention is a terminal capable of performing the functions mentioned in the present invention, for example, a wired / wireless telephone capable of network communication, a tablet PC (Tablet PC). , a laptop (Laptop), a smart phone (Smartphone), a personal digital assistant (Personal Digital Assistant), it is preferred to be understood as a terminal of a concept including any one of the mobile communication terminal (Mobile Communication Terminal).

As described above, although the embodiments have been described with reference to the limited embodiments and drawings, various modifications and variations are possible from the above description by those skilled in the art.

11 shows an example of an internal configuration of a computing device according to an embodiment of the present invention, and in the following description, descriptions of unnecessary embodiments overlapping with those of FIGS. 1 to 10 will be omitted. do it with

11, the computing device 10000 includes at least one processor 11100, a memory 11200, a peripheral interface 11300, an input/output subsystem ( It may include at least an I/O subsystem) 11400 , a power circuit 11500 , and a communication circuit 11600 . In this case, the computing device 10000 may correspond to a user terminal connected to the tactile interface device (A) or the aforementioned computing device (B).

The memory 11200 may include, for example, high-speed random access memory, magnetic disk, SRAM, DRAM, ROM, flash memory, or non-volatile memory. there is. The memory 11200 may include a software module, an instruction set, or other various data required for the operation of the computing device 10000 .

In this case, access to the memory 11200 from other components such as the processor 11100 or the peripheral device interface 11300 may be controlled by the processor 11100 .

Peripheral interface 11300 may couple input and/or output peripherals of computing device 10000 to processor 11100 and memory 11200 . The processor 11100 may execute a software module or an instruction set stored in the memory 11200 to perform various functions for the computing device 10000 and process data.

The input/output subsystem 11400 may couple various input/output peripherals to the peripheral interface 11300 . For example, the input/output subsystem 11400 may include a controller for coupling a peripheral device such as a monitor, keyboard, mouse, printer, or a touch screen or sensor as needed to the peripheral interface 11300 . According to another aspect, input/output peripherals may be coupled to peripheral interface 11300 without going through input/output subsystem 11400 .

The power circuit 11500 may supply power to all or some of the components of the terminal. For example, the power circuit 11500 may include a power management system, one or more power sources such as batteries or alternating current (AC), a charging system, a power failure detection circuit, a power converter or inverter, a power status indicator, or a power source. It may include any other components for creation, management, and distribution.

The communication circuit 11600 may enable communication with another computing device using at least one external port.

Alternatively, as described above, if necessary, the communication circuit 11600 may include an RF circuit to transmit and receive an RF signal, also known as an electromagnetic signal, to enable communication with other computing devices.

The embodiment of FIG. 11 is only an example of the computing device 10000, and the computing device 11000 omits some components shown in FIG. 11 or further includes additional components not shown in FIG. 11, or 2 It may have a configuration or arrangement that combines two or more components. For example, a computing device for a communication terminal in a mobile environment may further include a touch screen or a sensor in addition to the components shown in FIG. 11 , and various communication methods (WiFi, 3G, LTE) are provided in the communication circuit 1160 . , Bluetooth, NFC, Zigbee, etc.) may include a circuit for RF communication. Components that may be included in the computing device 10000 may be implemented in hardware, software, or a combination of both hardware and software including an integrated circuit specialized for one or more signal processing or applications.

Methods according to an embodiment of the present invention may be implemented in the form of program instructions that can be executed through various computing devices and recorded in a computer-readable medium. In particular, the program according to the present embodiment may be configured as a PC-based program or an application dedicated to a mobile terminal. The application to which the present invention is applied may be installed in the user terminal through a file provided by the file distribution system. For example, the file distribution system may include a file transmission unit (not shown) that transmits the file in response to a request from the user terminal.

The device described above may be implemented as a hardware component, a software component, and/or a combination of the hardware component and the software component. For example, the devices and components described in the embodiments may include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), It may be implemented using one or more general purpose or special purpose computers, such as a programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions. The processing device may run an operating system (OS) and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For convenience of understanding, although one processing device is sometimes described as being used, one of ordinary skill in the art will recognize that the processing device includes a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that can include For example, the processing device may include a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as parallel processors.

Software may comprise a computer program, code, instructions, or a combination of one or more thereof, which configures a processing device to operate as desired or is independently or collectively processed You can command the device. The software and/or data may be any kind of machine, component, physical device, virtual equipment, computer storage medium or device, to be interpreted by or to provide instructions or data to the processing device. may be permanently or temporarily embody in The software may be distributed over networked computing devices, and stored or executed in a distributed manner. Software and data may be stored in one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and used by those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described with reference to the limited embodiments and drawings, various modifications and variations are possible from the above description by those skilled in the art. For example, the described techniques are performed in an order different from the described method, and/or the described components of the system, structure, apparatus, circuit, etc. are combined or combined in a different form than the described method, or other components Or substituted or substituted by equivalents may achieve an appropriate result. Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the following claims.

Claims (10)

In the exercise posture correction method through the overlap analysis of images implemented in a computing device comprising one or more processors and a main memory for storing instructions executable by the processor,
An overlapping image generating step of generating an overlapping image by overlapping a first image input as a reference image for an exercise posture from a user terminal and a second image input as a comparison image with respect to the first image;
In the overlapping image, a region including at least one of a body joint region and an exercise tool region is an overlapping degree of movement postures of the first object included in the first image and the second object included in the second image an overlap error analysis step of analyzing an overlap error of the second object with respect to the first object by comparing the size and direction between images or objects by setting as an overlap reference area; and
A feedback information providing step of providing, to the user terminal, feedback information for minimizing an overlap error with respect to an exercise posture of the first object and the second object based on the result of the overlap error analysis;
Before performing the overlap error analysis step,
From the superimposed image, the frame-by-frame movement postures of the first object and the second object for the image section corresponding to the movement motion are extracted;
By applying the extracted movement postures for each frame to the deep learning algorithm, one or more key postures for movement movements are detected,
The overlapping error analysis step is
Acquire XY coordinate parameters for one overlapping reference area of the first object and the second object corresponding to the detected core posture change point, so that the trajectory change value of the movement motion for the one overlapping reference area is calculated and,
By comparing the trajectory change values calculated for the first object and the second object, the overlap error for the movement posture is analyzed,
The overlapping error analysis step is
Whenever an input to the second image is performed from the user terminal, the trajectory change value of the second object included in the second image is calculated, and from the trajectory change value of the second object accumulated in time order in the database Derive the average value and variance value for each core posture of the user, and analyze the change in posture stability of the user over time by analyzing the posture stability for each key posture of the user from the variance value,
The step of providing the feedback information,
Superimposition of images, characterized in that, through an external network, an image of an exercise posture of a professional player having a similarity between the average value for each core posture of the user and a preset threshold or more is searched, and the searched image is recommended as the first image. A method of correcting exercise posture through analysis.
The method of claim 1,
The step of generating the superimposed image includes:
The size and direction are corrected by setting an area including at least one of a body joint area and an exercise tool area of the first object included in the first image and the second object included in the second image as an overlapping reference area. By doing so, the exercise posture correction method through the overlap analysis of images, characterized in that generating an overlapping image in which the first image and the second image are combined.
delete delete delete delete The method of claim 1,
The overlapping error analysis step is
With respect to the detected core posture, the switching speed between the core postures included in the exercise motion is calculated using the time required to change the posture from the first core posture to the second core posture,
A method of correcting an exercise posture through an image overlap analysis, characterized in that the overlapping error for the exercise posture is analyzed by comparing the switching speed between the core postures calculated for the first object and the second object.
The method of claim 1,
The step of providing the feedback information,
When an image of a professional athlete's exercise posture is input as the first image,
As one of the feedback information provided to the user terminal, purchase link information for one or more exercise items extracted from the first image is provided.
In the exercise posture correction apparatus through superposition analysis of images implemented by a computing device including one or more processors and a main memory for storing instructions executable by the processor,
an overlapping image generating unit generating an overlapping image by superimposing a first image input as a reference image for an exercise posture from a user terminal and a second image input as a comparison image with respect to the first image;
In the overlapping image, a region including at least one of a body joint region and an exercise tool region is an overlapping degree of movement postures of the first object included in the first image and the second object included in the second image an overlap error analysis unit that analyzes an overlap error of the second object with respect to the first object by comparing the size and direction between images or objects by setting as an overlap reference region; and
A feedback information providing unit that provides feedback information to the user terminal for minimizing an overlap error with respect to an exercise posture of the first object and the second object based on the result of the overlap error analysis;
Before performing the function of the overlap error analysis unit,
From the superimposed image, the frame-by-frame movement postures of the first object and the second object for the image section corresponding to the movement motion are extracted;
By applying the extracted movement postures for each frame to the deep learning algorithm, one or more key postures for movement movements are detected,
The overlap error analysis unit,
Acquire XY coordinate parameters for one overlapping reference area of the first object and the second object corresponding to the detected core posture change point, so that the trajectory change value of the movement motion for the one overlapping reference area is calculated and,
By comparing the trajectory change values calculated for the first object and the second object, the overlap error for the movement posture is analyzed,
The overlap error analysis unit,
Whenever an input to the second image is performed from the user terminal, the trajectory change value of the second object included in the second image is calculated, and from the trajectory change value of the second object accumulated in time order in the database Derive the average value and variance value for each core posture of the user, and analyze the change in posture stability of the user over time by analyzing the posture stability for each key posture of the user from the variance value,
The feedback information providing unit,
Superimposition of images, characterized in that, through an external network, an image of an exercise posture of a professional player having a similarity between the average value for each core posture of the user and a preset threshold or more is searched, and the searched image is recommended as the first image. Exercise posture correction device through analysis.
A computer-readable recording medium comprising:
The computer-readable medium stores instructions for causing a computing device to perform the following steps:
an overlapping image generating step of generating an overlapping image by superimposing a first image input as a reference image for an exercise posture from a user terminal and a second image input as a comparison image with respect to the first image;
In the overlapping image, a degree of overlap between the motion postures of the first object included in the first image and the second object included in the second image is an area including at least one of a body joint area and an exercise tool area an overlap error analysis step of analyzing an overlap error of the second object with respect to the first object by comparing the size and direction between images or objects by setting as an overlap reference area; and
A feedback information providing step of providing, to the user terminal, feedback information for minimizing an overlap error with respect to an exercise posture of the first object and the second object based on the result of the overlap error analysis to the user terminal;
Before performing the overlap error analysis step,
From the superimposed image, the frame-by-frame movement postures of the first object and the second object for the image section corresponding to the movement motion are extracted;
By applying the extracted movement postures for each frame to the deep learning algorithm, one or more key postures for movement movements are detected,
The overlapping error analysis step is
Acquire XY coordinate parameters for one overlapping reference area of the first object and the second object corresponding to the detected core posture change point, so that the trajectory change value of the movement motion for the one overlapping reference area is calculated and,
By comparing the trajectory change values calculated for the first object and the second object, the overlap error for the movement posture is analyzed,
The overlapping error analysis step is
Whenever an input to the second image is performed from the user terminal, the trajectory change value of the second object included in the second image is calculated, and from the trajectory change value of the second object accumulated in time order in the database. Derive the average value and variance value for each core posture of the user, and analyze the change in posture stability of the user over time by analyzing the posture stability for each key posture of the user from the variance value,
The step of providing the feedback information,
Computer-reading, characterized in that, through an external network, an image of a professional athlete's exercise posture having a similarity between the average value for each core posture of the user and a preset threshold or higher is searched, and the retrieved image is recommended and provided as the first image possible recording medium.

Images