Practice support device, practice support method and program

Abstract

To provide a training support device, a training support method, and a program which enable a trainee to recognize, on a real-time basis, whether the training is carried out effectively.SOLUTION: A training support device (100) for supporting physical training includes a motion information acquisition device (20) that obtains motion information of a trainee, an error derivation part (12) for deriving an error between model motion and motion of the trainee obtained by the motion information acquisition device, and a stimulation generation part (14) for generating stimulation corresponding to the magnitude of the error to be provided for the trainee.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a practice support device, a practice support method, and a program. [Background technology]

[0002] Various training or rehabilitation methods have been proposed for people whose limb motor abilities have declined due to injury, aging, etc. For example, Patent Document 1 (JP Patent Publication No. 2017-196099) discloses an upper limb motor learning device that includes a virtual bone image generating means for generating a virtual bone image from a distance image captured by an imaging means, and an image generating means for displaying the virtual bone image so as to be superimposed on a teaching image. This device is said to enable appropriate rehabilitation to be performed by moving one's own limbs in accordance with the movements of the teaching image while viewing a composite image. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2017-196099 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, the above-mentioned conventional techniques have a problem in that the practitioner cannot obtain clues in real time as to whether or not the practice is being performed effectively. One aspect of the present invention aims to realize a practice support device, a practice support method, and a program that enable the practitioner to recognize in real time whether or not the practice is being performed effectively. [Means for solving the problem]

[0005] In order to solve the above problems, a practice support device according to one embodiment of the present invention is a practice support device that supports physical practice, and includes a movement information acquisition device that acquires movement information of a practitioner, an error derivation unit that derives an error between a model movement and the movement of the practitioner acquired from the movement information acquisition device, and a stimulus generation unit that generates a stimulus to be provided to the practitioner according to the magnitude of the error.

[0006] According to the practice support device having the above configuration, the practicer can recognize in real time whether or not the practice is being performed effectively.

[0007] In one embodiment of the practice support device of the present invention, the device further includes an overlay unit that overlays an image of the model action and an image of the practicer's action, and the stimulus generation unit generates an image as a visual stimulus to be provided to the practicer, and in the image in which the image of the model action and the image of the practicer's action are overlaid, the transparency of the image of the model action may be changed depending on the magnitude of the error.

[0008] According to the practice support device having the above configuration, when the difference between the actual movement and the model movement becomes large, the transparency of the image of the model movement becomes smaller, so that the practicer can recognize in real time whether the practice is being performed effectively or not. Furthermore, it becomes easier to imitate the model movement, and the effect of the practice can be improved.

[0009] In the practice support device according to one aspect of the present invention, the video of the model movement and the video of the movement of the practicer may be images seen from the line of sight of the practicer.

[0010] According to the practice support device having the above configuration, an image seen from the practitioner's line of sight is displayed, so that the practice can be performed with a natural impression.

[0011] In a practice support device according to one embodiment of the present invention, the error derivation unit may include a smoothing unit that smoothes a temporal change in the magnitude of the error, and the stimulus generation unit may generate a stimulus according to the magnitude of the smoothed error.

[0012] According to the practice support device having the above configuration, it is possible to smooth out fluctuations in the amount of error and reduce sudden changes in shading in the image of the model movement.

[0013] In the practice support device according to one aspect of the present invention, the stimulus generation section may generate a stimulus according to a magnitude of the error with a delay to a time at which the error is detected.

[0014] According to the practice support device having the above configuration, it is possible to convey to the practicer the effect of his or her own corrective action in an easily understandable manner.

[0015] In the practice support device according to one aspect of the present invention, the image may be a three-dimensional image.

[0016] According to the practice support device having the above configuration, the practicer feels as if he or she is looking at a real upper limb, making it easier for the practicer to practice.

[0017] In the practice support device according to one aspect of the present invention, the motion information acquisition device may include a sensor that detects the motion of the practicer.

[0018] According to the practice support device having the above configuration, the movement of the practicer can be detected by the sensor.

[0019] In a practice support device according to an aspect of the present invention, the motion information acquisition device may include an imaging device that images the motion of the practicer.

[0020] According to the practice support device having the above configuration, the movement of the practicer can be detected by the imaging device.

[0021] The practice assistance device according to an aspect of the present invention may further include a background generation unit that generates an environmental image that serves as a background.

[0022] According to the practice support device having the above configuration, it is possible to eliminate the problem of the practicer feeling unwell due to the lack of a background.

[0023] A practice support method according to one aspect of the present invention is a method for supporting a practice of moving a body, the method comprising the steps of: acquiring motion information of a practicer; deriving an error between a model motion and a motion of the practicer; and generating a stimulus according to a magnitude of the error. Includes.

[0024] According to the practice support method having the above configuration, the practicer can recognize in real time whether or not the practice is being performed effectively.

[0025] The practice support device of the present invention may be realized by a computer. In this case, the practice support program of the practice support device, which causes a computer to operate as each unit (software element) of the practice support device to realize the error derivation unit and the stimulus generation unit in a computer, and the computer-readable recording medium on which it is recorded, also fall within the scope of the present invention. Effect of the Invention

[0026] According to one aspect of the present invention, it is possible to realize a practice support device, a practice support method, and a program that allow a practicer to recognize in real time whether or not practice is being performed effectively. [Brief description of the drawings]

[0027] [Figure 1] 1 is a block diagram showing the configuration of a practice support device according to a first embodiment of the present invention. [Diagram 2] 1 is a perspective view showing the appearance of a display according to a first embodiment of the present invention. [Diagram 3] 4 is an example of a flowchart of a method for constructing a three-dimensional model of an exemplary movement and a method for constructing a three-dimensional model of a movement of a practicer according to the first embodiment. [Figure 4] 1 is an example of a method for supporting a practice of moving a body performed by a practicer according to the first embodiment, and a more detailed flowchart of the practice support method. [Diagram 5]1 is a graph showing images of a model movement and a movement of a trainee displayed on a display in accordance with the first embodiment, and a relationship between an amount of error and a transparency of the images of the model movement. [Figure 6] 13 is an example of a motion information acquiring device according to a modified example. [Figure 7] FIG. 5 is a block diagram showing the configuration of a practice support device according to a second embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0028] [Embodiment 1] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below. Fig. 1 is a block diagram of a practice support device 100 according to the first embodiment.

[0029] The practice support device 100 according to the first embodiment is a device that supports the practice of moving the body performed by, for example, a person who needs rehabilitation or the elderly. The practice support device 100 includes a control device 10. The practice support device 100 also includes a motion information acquisition device 20 and a display 30. The motion information acquisition device 20 acquires motion information of a trainee. A trainee is a person whose body function has been reduced due to injury, illness, aging, or the like, and who practices moving his or her body. The display 30 provides the trainee with a video of the state of his or her practice. The control device 10 derives an error from the model motion from the motion information of the trainee acquired from the motion information acquisition device 20, and causes the display 30 to display a video that serves as a stimulus according to the magnitude of the error. Each device will be described in detail below.

[0030] (Control device) The control device 10 includes a three-dimensional model construction unit 11, an error derivation unit 12, a superposition unit 13, a stimulus generation unit 14, a communication unit 15, and a storage unit 16. The control device 10 may be, for example, a personal computer such as a desktop computer or a notebook computer.

[0031] The three-dimensional model construction unit 11 analyzes the movement information of the trainee acquired by the movement information acquisition device 20, and constructs a three-dimensional model of the movement of the trainee. The movement information is information that can specify what kind of movement was made. The three-dimensional model is a model that displays the three-dimensional movement of each part of the body as an image. For example, the three-dimensional model is computer graphics that simulate each part of the body. The three-dimensional model construction unit 11 also generates a three-dimensional model of the model movement. The model movement is a movement that is considered preferable or a model when practicing moving the body. The model movement may be, for example, a movement performed by an occupational therapist, a physical therapist, or a fitness instructor, or may be a movement performed by other experts. In the following, an occupational therapist, a physical therapist, a fitness instructor, or other experts is called a model mover.

[0032] In the present embodiment 1, the three-dimensional model construction unit 11 generates an image of a model movement and an image of the movement of the practicer as seen from the practicer's line of sight. That is, the three-dimensional model construction unit 11 generates an image of the practicer's own movement as seen from the practicer's line of sight and an image of the model movement as seen from the same direction. Since the image of the model movement and the image of the movement of the practicer are images seen from the practicer's line of sight, the practicer can practice with a natural impression. Furthermore, it is preferable that this image is a three-dimensional image. Since the image is a three-dimensional image, the practicer feels as if he or she is looking at a real upper limb, making it easier for the practicer to practice. The method of constructing a model executed by the three-dimensional model construction unit 11 will be described later.

[0033] The error derivation unit 12 derives an error between a three-dimensional model of the movement of the trainee and a three-dimensional model of the model movement that was generated before the trainee started practicing. Specifically, the error is derived between a predetermined position of the three-dimensional model of the trainee and a predetermined position of the three-dimensional model of the model movement (a position corresponding to a predetermined position of the trainee's body). In this embodiment, the error is a distance in three-dimensional space. Specifically, the error is calculated by subtracting the three-dimensional position coordinates of the three-dimensional model of the trainee from the three-dimensional position coordinates of the model movement. The predetermined position for calculating the error can be, for example, a body position where an error is likely to occur, or a body position where it is easy to determine whether the movement is appropriate or not. Alternatively, instead of the error at a specific position, the error may be calculated at several points on the body and calculated by summing or averaging the errors.

[0034] The superimposing unit 13 generates an image by superimposing an image of a three-dimensional model of the model movement, which has been generated in advance, on an image of a three-dimensional model of the movement of the student.

[0035] The stimulus generating unit 14 generates a stimulus according to the magnitude of the error to be provided to the practitioner. In this embodiment, the stimulus refers to a clue that allows the practitioner to recognize the magnitude of the error (deviation) between the practitioner's action and the model action. In this embodiment, the stimulus generating unit 14 generates an image as a visual stimulus to be provided to the practitioner, and in an image in which an image of the model action and an image of the practitioner's action are superimposed, the transparency of the image of the model action is changed according to the magnitude of the error. Specifically, the smaller the error, the greater the transparency of the image of the model action, which the stimulus generating unit 14 increases.

[0036] More specifically, if the error is ε and the transparency of the image is Φ, the transparency Φ can be calculated, for example, using the following formula (1). Here, α and β are predetermined constants. It is preferable that α and β are determined in advance by experiments or the like so that they are appropriate transparency for allowing the trainee to recognize the magnitude of the error. Note that, since Φ increases as ε decreases, β is a negative number. Also, although formula (1) is a linear function (first-order formula), it is not limited to a first-order formula and may be another function formula in which the transparency decreases when the error is large.

number

[0037] As the error between the model movement and the movement of the trainee becomes smaller and the transparency increases, the image of the model movement becomes fainter and less noticeable. When the image of the model movement becomes fainter and less noticeable, the trainee can recognize that his or her movement is close to the model movement. Conversely, as the transparency of the image increases and the model movement becomes more noticeable, the trainee can recognize that his or her movement deviates significantly from the model movement.

[0038] The communication unit 15 performs information communication between the control device 10 and the motion information acquisition device 20. Specifically, the communication unit 15 receives sensor data from the motion information acquisition device 20. The communication unit 15 also performs information communication between the control device 10 and the display 30. Specifically, the communication unit 15 transmits video data to the display 30. The communication unit 15 may be, for example, a cable connector for wired communication, or an input / output interface for short-range wireless communication such as Bluetooth (registered trademark).

[0039] The storage unit 16 stores the constructed three-dimensional model. That is, the storage unit 16 stores a three-dimensional model of a model movement constructed in advance. The storage unit 16 also stores a three-dimensional model of a practicer constructed during the practice of the practicer.

[0040] (Operation information acquisition device) The motion information acquisition device 20 includes a marker 21, a sensor 22, and a communication unit 23. The marker 21 is an object to be detected by the sensor 22. The marker 21 is, for example, a reflector that reflects light from the sensor. The marker 21 may also be, for example, a light-emitting device such as an LED.

[0041] When practicing moving the upper limbs, the marker 21 is placed at at least one location, such as the wrist, elbow, etc. When practicing moving the lower limbs, the marker 21 is placed at at least one location, such as the ankle, knee, waist, etc.

[0042] The sensor 22 is, for example, a photosensor that can measure the spatial position (direction and distance) of the marker 21. For example, if the marker 21 is a reflector, the sensor 22 emits light and detects the reflected light from the reflector. The direction and distance to the reflector are calculated from the detected position of the reflected light and the time difference between the emission of light and the detection of the reflected light. If the marker 21 is a light-emitting device such as an LED, the sensor 22 is composed of two photodiodes. The direction and distance to the light-emitting device are calculated by trigonometry from the position of the light of the marker 21 detected by the two photodiodes. The three-dimensional model construction unit 11 can derive the spatial coordinates of the marker 21 from the detection data of the sensor 22.

[0043] The communication unit 23 transmits the signal acquired by the sensor 22 to the control device 10. The communication unit 23 may be, for example, a cable connector for wired communication, or may be an input / output interface for short-range wireless communication such as Bluetooth (registered trademark).

[0044] (display) Fig. 2 is a perspective view showing the appearance of the display 30 according to embodiment 1. As shown in Fig. 2, the display 30 takes the form of goggles (or a head mount) to be worn on the head of a practicer P. The display 30 includes therein a liquid crystal display unit for the right eye and a liquid crystal display unit for the left eye on which images are displayed.

[0045] An image in which the image of the model movement and the image of the movement of the learner are superimposed by the superimposing unit 13 is displayed on the display 30. Furthermore, an image in which the transparency of the model movement is changed by the stimulus generating unit 14 is displayed on the display 30.

[0046] The display 30 in the present embodiment 1 is attached to the head of the practitioner. Therefore, the display 30 can provide an image that allows the practitioner to see the movement of his / her own upper limbs from his / her own line of sight.

[0047] The form of the display 30 is not limited to the goggle type. For example, it may be a projection device that projects an image onto a screen or a wall. For example, when a practitioner practices full-body movements, the practitioner can practice while watching an image of his / her whole body projected onto a screen or the like. In this case, the projected image may be switched between an image seen from the front, an image seen from the side, an image seen from an oblique angle, or the like, or multiple images may be projected simultaneously.

[0048] The control device 10, the motion information acquisition device 20, and the display 30 of the practice support device 100 in the present embodiment 1 are configured as independent devices. However, some of the control device 10, the motion information acquisition device 20, and the display 30 may be integrated together. For example, the control device 10 and the motion information acquisition device 20 (excluding the marker 21) may be integrated together.

[0049] Furthermore, control device 10 may be located at a position separate from motion information acquisition device 20 and display 30. In that case, control device 10, motion information acquisition device 20, and display 30 may be connected, for example, via the Internet. With this configuration, one control device 10, motion information acquisition devices 20, and displays 30 are prepared at multiple locations, respectively, so that multiple practitioners can practice in parallel at each location.

[0050] (How to build a 3D model) Next, a method of constructing a model executed by the three-dimensional model construction unit 11 will be described with reference to the drawings. In the following, an example of an upper limb movement will be described, but the lower limb movement or the whole body movement may be used. FIG. 3 is an example of a flowchart of a method S1 for constructing a three-dimensional model of a model movement and a method S2 for constructing a three-dimensional model of a trainee's movement. First, the method S1 for constructing a three-dimensional model of a model movement will be described. First, in step S40, the three-dimensional model construction unit 11 acquires movement information of the model performer. Specifically, the three-dimensional model construction unit 11 acquires the model movement of each part of the body from a signal acquired by the sensor 22. For example, when acquiring the upper limb movement, the movement of the marker 21 arranged at at least one place such as the wrist or elbow of the model performer is acquired.

[0051] Since the structure of the upper limbs of humans is common, by measuring the length of the upper arm and forearm of the model performer in advance, the three-dimensional movement of the entire upper limbs can be obtained from the movement of one marker 21. In order to obtain the movement of the model performer more accurately, it is preferable to obtain the movement of two or more places. For example, by obtaining the movement using two markers 21 at the elbow and wrist, the movement of the upper limbs can be obtained more accurately than by obtaining the movement using one marker 21. In addition, since the movement of the shoulders changes the movement of the wrist and elbow, it is preferable to attach the marker 21 to the shoulder position to obtain the movement of the shoulders. If the movement of the shoulders can be detected, the movement of the upper limbs alone can be identified by subtracting the movement of the shoulders.

[0052] In this way, data on the three-dimensional positions (spatial coordinates) of the elbow and wrist during the model movement and their movements can be obtained.

[0053] Next, in step S42, the three-dimensional model construction unit 11 constructs a three-dimensional model of the model movement. That is, a data set of how the upper limbs move in the model movement is constructed. The three-dimensional model is preferably a three-dimensional model that imitates the shape of a human upper limb. In the case of a three-dimensional model that imitates the shape of a human upper limb, the trainee feels as if he or she is looking at a real upper limb, which makes training easier.

[0054] Next, in step S44, the 3D model construction unit 11 records the constructed 3D model of the model movement in the storage unit 16. This completes the construction of the 3D model of the model movement. Note that it is preferable that the 3D model of the model movement is constructed before the practicer starts practicing.

[0055] Next, a method S2 for constructing a three-dimensional model of a movement of a practitioner will be described. First, in step S50, the three-dimensional model construction unit 11 acquires movement information of the practitioner. Specifically, the three-dimensional model construction unit 11 acquires the movement of each part of the body of the practitioner from a signal acquired by the sensor 22. For example, the movement of the marker 21 placed at least at one place such as the wrist or elbow of the practitioner is acquired. As described above, by measuring the length of each part of the arm of the practitioner in advance, the three-dimensional movement of the entire upper limb can be acquired from the movement of the marker 21 at one place. It is more preferable to acquire the movement using the markers 21 at two or more places. It is also preferable to attach the marker 21 to the shoulder position to acquire the shoulder movement.

[0056] Next, in step S52, the three-dimensional model constructing section 11 constructs a three-dimensional model of the movement of the practitioner. It is preferable that the three-dimensional model is a three-dimensional model that imitates the form of the upper limbs of the practitioner.

[0057] Next, in step S54, the 3D model construction unit 11 records the constructed 3D model of the movement of the practicer in the storage unit 16. This completes the construction of the 3D model of the movement of the practicer. The constructed 3D model of the movement of the practicer is transmitted to the display 30 in real time and displayed.

[0058] (Practice support method) Next, a method for supporting a practice in which a practitioner moves his / her body will be described. Fig. 4 shows an example of a flowchart of a practice support method S3 for supporting a practice in which a practitioner moves his / her body, and a more detailed practice support method S4. First, the practice support method S3 will be described.

[0059] First, in step S60, the three-dimensional model construction unit 11 acquires movement information of the practicer. This is the same as step S50 described with reference to FIG.

[0060] Next, in step S62, the error derivation unit 12 derives an error between the model motion and the motion of the practicer. Specifically, the error derivation unit 12 derives a spatial distance between the model motion and the elbow and wrist position coordinates from the model motion and the elbow and wrist position coordinates from the model motion acquired by the 3D model construction unit 11. Furthermore, the error derivation unit 12 may further derive an error in a predetermined position of the upper arm or the forearm.

[0061] Next, in step S64, the magnitude of the stimulus is changed according to the magnitude of the error. That is, the stimulus generator 14 generates a stimulus according to the magnitude of the error derived in step S62. This ends the method S3.

[0062] Next, the practice support method S4 will be described in more detail. First, in step S70, the three-dimensional model construction unit 11 acquires the movement information of the practicer. Next, in step S72, the error derivation unit 12 derives the error between the model movement and the movement of the practicer. Steps S70 and S72 are similar to steps S60 and S62.

[0063] Next, in step S74, the superimposing unit 13 superimposes the image of the model movement and the image of the movement of the trainee on the display 30. Each of the two images is a three-dimensional model image. For example, in the case of an upper limb movement, the superimposing unit 13 preferably superimposes the shoulder positions of the model movement and the shoulder positions of the trainee at the same positions. The image of the three-dimensional model of the model movement is stored in the storage unit 16 in advance.

[0064] Next, in step S76, the stimulus generation unit 14 changes the transparency of the video of the model behavior displayed on the display 30 according to the magnitude of the error derived in step S72. Specifically, the stimulus generation unit 14 reduces the transparency of the video of the model behavior as the error amount (magnitude of the error) increases (see FIG. 5). The video of the model behavior with its transparency changed is transmitted to the display 30 and displayed.

[0065] The above-mentioned steps S72 to S76 are executed sequentially from the time when the movement of the practicer is detected in step S70. In other words, the practicer can see that the transparency of the image of the model movement changes in accordance with his / her own movement at almost the same timing as his / her own movement.

[0066] This allows the student to recognize in real time that the more the image of the model movement becomes darker, the more the student's own movement deviates from the model movement, and the student can attempt to make his or her own movement overlap with the image of the model movement.

[0067] 501 in FIG. 5 is an example of an image of a three-dimensional model of the model movement (shown by dotted lines) and a three-dimensional model of the movement of a practitioner (shown by solid lines) displayed on the display 30. The closer the upper limbs of the practitioner are to the upper limbs of the model movement, the greater the transparency of the image of the three-dimensional model of the model movement, that is, the lighter the image becomes. 502 in FIG. 5 is a graph showing the relationship between the amount of error and the transparency of the image of the model movement. As shown in this graph, the greater the error, the smaller the transparency.

[0068] As described above, the practice support device 100 and practice support methods S3 and S4 according to the first embodiment allow the practicer to recognize in real time whether his / her practice is being performed effectively. Specifically, by changing the transparency of the model action displayed on the display 30, the practicer can recognize in real time whether his / her movement is close to or different from the model action. Furthermore, it becomes easier to imitate the model action, and the effect of practice can be improved. Therefore, the effect of practice is greater than if the practicer were to practice without knowing whether his / her movement is close to the model action.

[0069] (Variation 1) In the above embodiment, the stimulus generating unit 14 changes the transparency of the image of the model movement. In other words, the transparency of the image becomes the size of the stimulus. This allows the trainee to recognize the size of the error (deviation) between his / her own movement and the model movement in real time.

[0070] However, the type of the stimulus is not limited to this. For example, the stimulus provided to the practicer may be vibration given to the practicer. In this case, a vibrator is attached to the upper or lower limb of the practicer. Then, the greater the error, the greater the signal the stimulus generating unit 14 generates to cause the vibrator to generate a larger vibration. The stimulus provided to the practicer may also be sound. A speaker or alarm is placed in a position where the practicer can hear it, and the greater the error, the greater the signal the stimulus generating unit 14 generates to cause a larger sound. Even with such a stimulus, the practicer can be informed in real time of the magnitude of the error (deviation) between his / her own movement and the model movement. In this case, it is not necessary to present an image to the practicer, but the effect of the practice will be higher if an image is present.

[0071] (Variation 2) In the above embodiment, the motion information acquisition device 20 includes a marker 21 and a sensor 22. However, the motion information acquisition device 20 may have any configuration that can acquire the motion information of the practitioner, and the configuration is not limited. For example, as shown in FIG. 6, the motion information acquisition device 20 may be a motion information acquisition device 20A that includes an imaging device 24 and a communication unit 25. The imaging device 24 captures the motion of the practitioner and acquires position data of each part of the body of the practitioner. The method of acquiring the position data of each part of the body of the practitioner may be to wear a marker and detect its position. Alternatively, the imaging device may be placed at a position where it can be captured by the line of sight of the practitioner, and the three-dimensional model construction unit 11 may generate a three-dimensional model of the practitioner based on the image.

[0072] The imaging devices 24 may be disposed one on each side of the goggle-type display 30. This allows an image for the right eye to be acquired from the imaging device 24 on the right side, and an image for the left eye to be acquired from the imaging device 24 on the left side, and these images can be displayed on the liquid crystal display section for the right eye and the liquid crystal display section for the left eye of the display 30, respectively, to display a stereoscopic image.

[0073] Furthermore, the motion information acquisition device 20 may be a motion information acquisition device 20B equipped with a sensor 26 that directly detects the motion of the practicer without using a marker, and a communication unit 27. The sensor 26 is, for example, an acceleration sensor. The acceleration sensor 26 may be attached to, for example, the wrist of the practicer, and position information of the wrist, elbow, etc. may be derived from the movement direction and movement speed of the upper limb from the output of the acceleration sensor 26.

[0074] [Embodiment 2] Other embodiments of the present invention will be described below with reference to the drawings. For ease of explanation, the same reference numerals are given to members having the same functions as those described in the first embodiment, and the description thereof will not be repeated.

[0075] 7 is a block diagram of a practice support device 100A according to embodiment 2. The practice support device 100A differs from the practice support device 100 according to embodiment 1 in that the error derivation unit 12 includes a smoothing unit 121.

[0076] The smoothing unit 121 smoothes the temporal change in the magnitude of the error. Specifically, the smoothing unit 121 calculates a time moving average of the magnitude of the error derived by the error derivation unit 12 and outputs it as an amount of error. The stimulus generation unit 14 calculates a transparency according to the amount of error.

[0077] The smoothing unit 121 calculates the amount of error of the time moving average as follows. First, the smoothing unit 121 acquires, for example, the following formula (2) which is a time series of the amount of error ε at measurement time i every 10 ms calculated by the error derivation unit 12.

[0078]

number

[0079] Next, the smoothing unit 121 calculates a moving average ma of the error amount ε for a section number of 51 frames (0.51 seconds assuming 10 ms per frame) according to the following formula (3).

[0080]

number

[0081] Next, the stimulus generating unit 14 calculates the transparency Φ from the moving average ma of the calculated error amount ε according to the following formula (4). Here, α and β are predetermined constants. In addition, β is a negative number so that the transparency is reduced as the error amount increases. It is preferable that α and β are determined in advance by experiments or the like so that they are appropriate transparency for allowing the practitioner to recognize the error amount.

[0082]

number

[0083] As described above, the transparency Φ can be expressed by a linear function including a moving average of the error amount ε between the model movement and the 3D model of the practicer. Note that the transparency Φ in the graph shown in 502 of FIG. 5 is calculated as described above.

[0084] If the error amount is calculated every short time, the change in the error amount becomes large. If the transparency is changed according to the error amount, the change in transparency also becomes large, so that the image of the displayed model movement may suddenly become lighter or darker, which may make it difficult for the trainee to see. However, by smoothing the variation in the error amount using the smoothing unit 121, it is possible to reduce the sudden change in the density of the image of the model movement.

[0085] In the above-mentioned formula (2), the measurement time is set to every 10 ms, and the moving average ma of the error amount ε in formula (3) is calculated over 51 frames (1 frame = 10 ms) in an example. However, the measurement time and the number of frames are not limited to this, and can be set arbitrarily within a range in which smoothing is performed appropriately. In addition, formula (4) is a linear function (first-order expression), but it is not limited to a first-order expression, and other function expressions that reduce the transparency when the error amount is large may be used.

[0086] Furthermore, by taking a time-moving average of the error amount, the stimulus generating unit 14 can generate a stimulus according to the magnitude of the error, delayed from the time when the error was detected. That is, the timing of the change in the transparency of the displayed image can be delayed. This provides the following advantages. That is, if the process of changing the transparency according to the magnitude of the error is fast, the change in transparency in response to the corrective action of the trainee may be too sudden, and the trainee may not be able to keep up. That is, if the transparency changes at the moment the action is corrected, it becomes difficult to determine whether the change in transparency is due to the trainee's own correction of the action or the result of the trainee's own action before the correction. Therefore, changing the transparency at a timing slightly delayed from the trainee's corrective action makes it easier to convey to the trainee the effect of his or her own corrective action. By taking a time-moving average of the error amount, such advantages can also be obtained.

[0087] The timing of the change in transparency in response to the error amount can be delayed by increasing the time for taking the moving average. Therefore, by adjusting the time for taking the moving average, the delay time can be adjusted so that the practicer's sense of correction is appropriately reflected.

[0088] The method of delaying the timing of the change in transparency relative to the amount of error is not limited to taking a moving average of the error. For example, the transparency may be changed by simply delaying the error by a predetermined time from the time the error was acquired, and then displayed on the display 30.

[0089] (Variation 3) As shown in FIG. 7, the practice support device 100A may include an environmental image capturing device 40, and the control device 10A may include a background generating section 17.

[0090] The background generation unit 17 generates a background environmental image based on an image acquired by the environmental image acquisition device 40. The environmental image is an image of the scenery of the place where the practicer is practicing. The environmental image acquisition device 40 is attached to the side of the display 30. Therefore, the environmental image acquisition device 40 can acquire a background image seen from the practicer's perspective. The environmental image acquisition device 40 may also serve as the imaging device 24 of the motion information acquisition device 20A described in the second modification.

[0091] The background image acquired by the environmental image acquisition device 40 is sent to the control device 10A, and is generated as a computer graphic environmental image by the background generation unit 17. This environmental image is further composited as a background onto the image of the model movement and the three-dimensional model of the practicer by, for example, the superimposition unit 13. Furthermore, the storage unit 16 may store the environmental image generated by the background generation unit 17, and may composite the stored environmental image without using the environmental image acquisition device 40 during the next practice.

[0092] If the images displayed on the display 30 were only the images of the trainee's movements and the model movements, with no background, the trainee might lose track of where he or she is and feel unwell. However, by also displaying an image of the background environment, this can be prevented.

[0093] In the above-described embodiment 2, the same effects as those obtained in embodiment 1 can be obtained. Note that modified examples 1 and 2 described in embodiment 1 can also be applied to embodiment 2, and the respective effects can be obtained. Modified example 3 described in embodiment 2 can also be applied to embodiment 1, and the same effect can be obtained.

[0094] [Software implementation example] The control device 10 of the practice support device 100 (in particular, the 3D model construction unit 11, the error derivation unit 12, the superposition unit 13, and the stimulus generation unit 14) may be realized by a logic circuit (hardware) formed in an integrated circuit (IC chip) or the like, or may be realized by software.

[0095] In the latter case, the practice support device 100 includes a computer that executes instructions of a program, which is software that realizes each function. The computer includes, for example, one or more processors, and a computer-readable recording medium that stores the program. The computer achieves the object of the present invention by having the processor read the program from the recording medium and execute it. The processor may be, for example, a CPU (Central Processing Unit). The recording medium may be a "non-transient tangible medium," such as a ROM (Read Only Memory), tape, disk, card, semiconductor memory, programmable logic circuit, or the like. The device may further include a RAM (Random Access Memory) that expands the program. The program may be supplied to the computer via any transmission medium (such as a communication network or broadcast waves) that can transmit the program. One aspect of the present invention may also be realized in the form of a data signal embedded in a carrier wave, in which the program is embodied by electronic transmission.

[0096] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. The technical scope of the present invention also includes embodiments obtained by appropriately combining the technical means disclosed in the different embodiments. Furthermore, new technical features can be formed by combining the technical means disclosed in the respective embodiments. [Explanation of symbols]

[0097] 10, 10A Control Device 11 3D Model Construction Section 12 Error derivation section 121 Smoothing section 13 Overlapped section 14 Stimulus generation section 15, 23, 25, 27 Communications Department 16 Memory section 17 Background generation section 20, 20A, 20B Operation information acquisition device 21 Marker 22 Sensors 24 Imaging device 26 Sensors 30 Display 40 Environmental image acquisition device 100, 100A Practice Support Device

Claims

1. A training support device that supports physical training, A motion information acquisition device that acquires motion information representing a motion of a practicer; an error derivation unit that derives a smoothed error by smoothing a temporal change in the magnitude of an error between the model motion and the motion of the practicer; a stimulus generator that generates a stimulus corresponding to the magnitude of the error to be provided to the practicer; a superimposing unit for superimposing an image of the model movement and an image of the movement of the trainee, the stimulus generation unit reduces a transparency of the image of the model movement as the smoothing error increases in an image in which the image of the model movement and the image of the movement of the learner are superimposed on each other; Practice support device.

2. 2. The training support device according to claim 1, wherein the video of the model movement and the video of the movement of the trainee are images seen from the trainee's line of sight.

3. The practice support device according to claim 1 , wherein the stimulus generating section generates the stimulus according to the magnitude of the error with a delay from a time at which the error is detected.

4. The training support device according to claim 1 , wherein the image is a three-dimensional image.

5. The practice support device according to claim 1 , wherein the motion information acquisition device includes a sensor that detects the motion of the practicer.

6. The practice support device according to claim 1 , wherein the motion information acquisition device includes an imaging device that images the motion of the practicer.

7. The practice support device according to claim 1 , further comprising a background generation unit that generates a background environmental image.

8. 1. A method for supporting physical activity training, comprising: One or more processors acquiring motion information representing a motion of a practice person; deriving a smoothed error by smoothing a temporal change in the magnitude of an error between the model movement and the movement of the practicer; generating a stimulus responsive to the magnitude of the error; a superimposing step of superimposing the image of the model movement and the image of the movement of the learner; In the step of generating the stimulus, in an image in which the image of the model movement and the image of the movement of the learner are superimposed, the greater the smoothing error, the smaller the transparency of the image of the model movement. Practice support method.

9. A practice support program for causing a computer to function as the practice support device according to any one of claims 1 to 7, the practice support program causing a computer to function as the error derivation unit, the stimulus generation unit, and the superposition unit.

Images