JP2014133015A - Body motion learning support device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a body motion learning support device that performs an evaluation of at least one evaluation item for each part divided on a time axis of a series of body motions, and performs an evaluation of at least one evaluation item for the entire series of body motions.SOLUTION: The body motion learning support device includes: a motion capture device 11 for acquiring body motion data of a body motion; a motion data storage device 12 for storing the body motion data and also storing model motion data used for comparison with the body motion data; an evaluation device 13 for evaluating the body motion using a score or a rank by comparing the body motion data and the model motion data; and an evaluation result display device 14 for presenting the body motion by an animation using the body motion data and the model motion data, and displaying the result of the evaluation by the evaluation device 13, and performs an evaluation of at least one evaluation item for each part divided on a time axis of a series of body motions, and also an evaluation for the entire series of body motions.

Description

  The present invention relates to a body motion learning support apparatus that evaluates a body motion acquired by practice and expresses the body motion in a form (for example, skeleton) so that a difference from the model motion can be understood. It is related with the physical movement learning assistance apparatus which can evaluate the physical motion of the learner acquired by the motion capture device, and can give an advice to the learner.

Conventionally, a dance learning support device using motion capture is known as a physical motion learning support device (see Non-Patent Document 1, Non-Patent Document 2, etc.).
For example, there is a Japanese dance learning support device that can recognize the difference between a skilled person and a learner by visually presenting the center of gravity of the body, which has been considered important in traditional Japanese dance movements. ing.
In addition, samba learning support that enables feedback on the difference between skilled and learners by visually presenting the temporal changes in the acceleration of body movements regarding the rhythmic characteristics that are considered important in samba movement. Devices are also known.
In these learning support devices, only a specific single evaluation item is evaluated for the entire series of body movements.

School Technology IEICE Technical Report MVE 2006-82 (2007-3) The Institute of Electronics, Information and Communication Engineers School Technique IEICE Technical Report CQ2008-67 (2009-01) The Institute of Electronics, Information and Communication Engineers

For example, in Japanese dance, a large number of sequences including a relatively large variety of basic motions often occur in choreography.
However, as described above, the conventional learning apparatus evaluates only a specific single evaluation item for the entire series of body movements, and thus is not suitable for learning Japanese dance as described above.
An object of the present invention is to evaluate at least one evaluation item for each part of a series of body movements divided on the time axis, or at least one evaluation item for a whole series of body movements. Is to evaluate.

The gist of the present invention is as follows.
(1)
In the body movement learning support device that evaluates the body movement of one or more learners acquired by practice and expresses the body movement by animation so that the difference from the model movement can be understood,
A motion capture device for acquiring the physical motion as physical motion data;
A motion data storage device that stores physical motion data acquired by the motion capture device and stores the exemplary motion data used for comparison with the physical motion data;
An evaluation device that evaluates the physical motion using a score or rank by comparing the physical motion data stored in the motion data storage device and the exemplary motion data;
An evaluation result display device for expressing the physical motion as the animation using the physical motion data and the exemplary motion data stored in the motion data storage device, and displaying the result of the evaluation by the evaluation device; ,
With
The evaluation apparatus evaluates at least one evaluation item for each part divided on the time axis for the series of body movements, or further, at least one evaluation item for the whole series of body movements Evaluate
A body movement learning support device characterized by the above.
(2)
The combination of the evaluation items is different for each of the parts divided on the time axis, the physical motion learning support device according to (1).
(3)
The physical motion learning support device according to (1), wherein the part and / or the evaluation item divided on the time axis are determined by the learner.
(4)
The physical motion learning support device according to any one of (1) to (3), wherein the physical motion and the exemplary motion are dances.

In the present invention, the “body movement” is an action in which a certain value is recognized in an action having a certain style, and is typically “dance” or “dance choreography” (choreography), for example, dance on ice (Figure skating) is also included. The dance may be an ethnic dance (for example, a Bon dance in Japan) or a social dance.
There may be one or more “learners”. When the “learner” is one person, the “exemplary movement” of one exemplary person corresponds to the “body movement”.
When there are multiple “learners”, the “body movement” can be an action that integrates the actions of multiple learners, and the “exemplary action” in this case is an action that integrates the actions of multiple learners. be able to.
In addition, when there are multiple “learners”, the “body movement” can be a plurality of body movements in which the movements of multiple learners are defined separately. It is a separate operation by corresponding multiple examples.
The body motion learning support device can express body motion by animation. This animation can usually be represented by a skeleton.
The motion capture device may be of any means as long as it can capture body motion, such as an optical detection type or a magnetic detection type. A motion capture device normally captures body movements by recognizing joint and head movements.
In the present invention, various evaluation items can be defined, such as matching of timings of predetermined parts of the body, smoothness of movement, and the like. The evaluation item can include a synchronous evaluation of motion and music. In addition, it is possible to include a match evaluation of the utterance timing of the shout.
It is preferable to use a plurality of evaluation items as the evaluation items. Preferable evaluation items for “dance” or “dance choreography” include “size of movement”, “key pose”, “choreography”, “cooperation”, “timing”, “smoothness”, Of these items, at least two, three, or five or more evaluation items may be used.
The evaluation device can express the evaluation as a score or a rank. Specifically, radar chart evaluation, ABC evaluation, and score evaluation are exemplified as the evaluation device. Statistical analysis of the results of scoring by seeing a large number of performances by professionals can make evaluation of specific actions meaningful.
The evaluation result display device expresses the body motion by animation and displays the result of evaluation by the evaluation device. The evaluation result display device can also display the evaluation result in an animation. For example, it is possible to divide the evaluation into three stages, and display the animation in blue for high evaluation, display the animation in yellow for middle evaluation, and display the animation in red for low evaluation. . For example, separate evaluations can be made for the upper and lower bodies, and the color of the evaluation can be changed for the upper and lower bodies.
The evaluation result display device can display a time axis that divides a series of body movements for each part. On this time axis, the evaluation for each divided part can be displayed on the time axis. For example, the evaluation can be displayed in different colors on the time axis, and the evaluation is divided into three stages. For a highly evaluated part, the part of the part on the time axis is displayed in blue, and for the middle evaluation part Can be displayed in yellow, and low-rated parts can be displayed in red. This makes it easier to see at a glance which part of the whole body movement needs to be improved. Also, when a specific part on the displayed time axis is selected, the animation of that part can be displayed.
In the present invention, the learner can be configured to be able to define a part divided on the time axis. For example, the learner can divide the total dance time into 12 or 16 segments. In addition, the learner can set the evaluation items as appropriate. For example, it is possible to have three evaluation items for a certain part and four evaluation items for another certain part. In this case, all or some of the evaluation items may or may not match.
In the present invention, the body motion of slow motion can be evaluated. In this case, the data to be compared with the slow motion body motion data is the slow motion model motion data.

Since the learner receives the evaluation of at least one evaluation item for each part divided on the time axis in the series of body movements, the learner can learn by feedback.
Further, the learner can receive a total evaluation of at least one evaluation item or a comprehensive evaluation of all the evaluation items with respect to the entire series of body movements, and learn by feedback.
By providing feedback, the degree of achievement and problems in each part are clarified, so that an excellent learning effect can be obtained.

FIG. 1 is a configuration block diagram showing an embodiment of a physical motion learning support apparatus of the present invention. FIG. 2 is a configuration diagram illustrating an example of the motion capture device illustrated in FIG. 1. FIG. 3 is a configuration diagram illustrating a process of evaluating a plurality of evaluation indexes using the exemplary operation data in the evaluation apparatus of FIG. FIG. 4 is a block diagram showing a process of dividing exemplary motion data into time segments based on choreographic division time data in the time segmentation device of FIG. FIG. 5 is a block diagram of a hierarchical neural network for deriving a comprehensive evaluation score by the evaluation apparatus of FIG. FIG. 6 is a diagram showing a display screen by the evaluation result display device.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a configuration block diagram showing an embodiment of the present invention.
The body motion learning support device 1 (body motion learning support device) in FIG. 1 evaluates the body motion acquired by practice, and in one form (for example, skeleton) the body motion so that the difference from the model motion can be understood. Can be expressed.

In FIG. 1, the physical motion learning support device 1 includes a motion capture device 11, a motion data storage device 12, an evaluation device 13, and an evaluation result display device 14.
The motion capture device 11 acquires the physical motion of the learner A as physical motion data DATA _MOT .
The motion data storage device 12 stores physical motion data DATA _MOT acquired by the motion capture device 11, and also stores exemplary motion data DATA _MDL used for comparison with the physical motion data DATA _MOT . In general, the model motion data DATA _MDL is obtained from the physical motion of a skilled person.

The evaluation device 13 evaluates the physical motion of the learner A using the score or rank by comparing the physical motion data DATA _MOT and the exemplary motion data DATA _MDL stored in the motion data storage device 12. The evaluation device 13 includes an item evaluation unit 131 and a comprehensive evaluation unit 132.
The item evaluation unit 131 compares the body motion data DATA _MOT and the model motion data DATA _MDL for each part divided on the time axis in a series of body motions, and evaluates a plurality of evaluation items. In this embodiment, although not shown in the drawing, the k-th evaluation item of the j-th part is represented by ITM _EST (j, k), and the evaluation value (score or rank) is represented by VL _EST (j, k). However, j = 1, 2, 3,..., M, and k = 1, 2, 3,.
The comprehensive evaluation unit 132 performs a total evaluation for each evaluation item or a single comprehensive evaluation on the entire series of body movements.
The total evaluation for each evaluation item is, for example, an evaluation value VL _EST (j, x) for a specific evaluation item ITM _EST (j, x) (x is any one of 1, 2, 3,..., N). ) For all parts.
In the single comprehensive evaluation, for example, an artificial intelligence type algorithm based on machine learning (in this embodiment, a neural network) is applied to the evaluation value VL _EST (j, x), and a total evaluation score TOTAL _EST of 100 points is obtained. Is generated.

The evaluation result display device 14 uses the body motion data DATA _MOT and the model motion data DATA _MDL stored in the motion data storage device 12 to express the body motion in a single form (in this embodiment, a skeleton animation). The result of evaluation by the evaluation device 13 is displayed.

FIG. 2 is a configuration diagram illustrating an example of the motion capture device 11.
In FIG. 2, the motion capture device 11 includes a magnetic sensor group 111 and a computer 112 attached to a learner's body.
Head, chest, waist, upper right arm, right forearm, right hand, left upper arm, left forearm, left hand, right thigh, right thigh, right foot, output from magnetic sensor group 111, The posture data of the left thigh, the left shin, the left foot, and the position data of the lumbar region are obtained by applying the human form (skeletal structure human body model MDL _SKLTN ) reproduced on the software in the computer 112, Neck joint, chest-lumbar joint, right shoulder joint, right elbow joint, right wrist joint, right hand tip, left shoulder joint, left elbow joint, left wrist joint, left hand tip, right hip joint, right knee joint, right ankle joint The time-series data obtained by acquiring the positions of the right toe part, the left hip joint, the left knee joint, the left ankle joint, and the left toe part at regular time intervals are output as the body motion data DATA _MOT .

FIG. 3 is a configuration diagram illustrating a process in which the item evaluation unit 131 evaluates a plurality of evaluation indexes using the exemplary operation data DATA _MDL .
The item evaluation unit 131 includes a time segmentation device 1311, a model data / learner data time difference calculation device 1312, and an item evaluation device 1313.
In FIG. 3, the time segmentation device 1311 performs choreography by the dance instructor as shown in FIG. 4 (a configuration diagram showing a process of dividing the exemplary motion data DATA _MDL into time segments based on the choreography division time data DATA _DIV ). According to the N divided times input as the divided time data DATA _DIV , the first time segment is from the start of operation to the first divided time of the choreographic divided time data DATA _DIV , and the i-th time segment (2 ≦ i ≦ N) is the ( i-1) The exemplary motion data DATA _MDL is divided into time segments from immediately after the end time of the time segment to the i-th division time, and the (N + 1) -th time segment is from immediately after the end of the N-th time segment to the end of operation. Output as data DATA _PART .

In FIG. 3, the time difference calculation device 1312 between model data / learner data calculates the right shoulder joint, right calculated from each joint position at each data acquisition time of the model motion data DATA _MDL regarding the learner's body motion data DATA _MOT. DP matching of posture generation time corresponding to the body posture of the whole body indicated by the angle of elbow joint, right wrist joint, left shoulder joint, left elbow joint, left wrist joint, right hip joint, right knee joint, left hip joint, left knee joint The difference of the generation time in the body movement data DATA _{MOT with} respect to the generation time in the model movement data DATA _MDL is output as the whole body posture time difference data DATA _{DIF_TB} .

The body posture of the upper body indicated by the angle of the right elbow joint, right wrist joint, left shoulder joint, left elbow joint, and left wrist joint at each data acquisition time of the model motion data DATA _MDL with respect to the learner's physical motion data DATA _{MOT Is} generated by the DP matching method, and the difference in the generation time in the body motion data DATA _{MOT with} respect to the generation time in the exemplary motion data DATA _MDL is output as the upper body posture time difference data DATA _{DIF_UB} .
Furthermore, regarding the learner's body movement data DATA _MOT , the lower body posture corresponding to the body posture of the lower body indicated by the angle of the right hip joint, right knee joint, left hip joint, and left knee joint at each data acquisition time of the model motion data DATA _{MDL Is} generated by the DP matching method, and the difference between the generation time in the body movement data DATA _{MOT and} the generation time in the model movement data DATA _MDL is output as lower body posture time difference data DATA _{DIF_LB} .

Evaluation items to be evaluated by the item evaluation device 1313 are selected in advance as follows.
A sample group of physical motion data DATA _MOT of a plurality of dance learners having different mastery degrees depending on the motion capture device 11 of FIG. 2 is acquired in advance, and animation is performed by using existing computer graphics animation production software using the physical motion data DATA _MOT. Create a group and write a free text group that the dance instructor evaluates the behavior of each animation in the animation group. The dance instructor evaluates the behavioral aspects by coding the sentence group using the SCAT method (Steps for Coding and Theorization method), which is a qualitative data analysis method, and categorizing the extracted code group for each evaluation index. Select multiple evaluation indicators used to do this. The item evaluation device 1313 outputs, for each time segment, an evaluation value VL _EST of the magnitude, key pose, choreography, cooperation, timing, and smoothness, which are items selected in the above procedure.

With respect to the evaluation value VL _EST , the evaluation value of the magnitude of movement is calculated as follows by separating the movement of the upper body and the movement of the lower body. During the i-th time segment, the trajectory followed by each of the seven points of the top of the exemplary motion data DATA _MDL , the right elbow joint, the right wrist joint, the right hand tip, the left elbow joint, the left wrist joint, and the left hand tip The length is derived, and the average value L _{T, U} (i) of the trajectory length is obtained. Similarly, the length of the trajectory traced by each of the seven points of the top, right elbow, right wrist, right hand, left elbow, left wrist, and left hand of the learner's physical movement data DATA _MOT The average value L _{S, U} (i) of the trajectory length is derived. L _T, L _S for U _(i), the difference _{L U (i) = L S} , U (i) -L T, upper body movement learner U (i) is in the i time segment of U (i) As an evaluation value VL _EST with 0 as an ideal value, it is assumed that it indicates the magnitude of the evaluation item movement with respect to.

On the other hand, during the i-th time segment, the length of the trajectory traced by each of the six points of the right knee joint, right ankle joint, right toe part, left knee joint, left ankle joint, and left toe part of the exemplary motion data DATA _MDL The average value L _{T, L} (i) of the length of the trajectory is obtained. Similarly, the length of the trajectory followed by each of the six points of the right knee joint, right ankle joint, right toe, left knee joint, left ankle joint, and left toe of the body motion data DATA _MOT is derived. An average value L _{S, L} (i) is obtained. L _T, L for L (i) _S, the difference _{L L (i) = L S} , L (i) -L T, lower body movements learner L a (i) in the segment i-th time L (i) As an evaluation value VL _EST with 0 as an ideal value, it is assumed that it indicates the magnitude of the evaluation item movement with respect to.

For the evaluation value VL _EST , the evaluation value of the key pose is calculated as follows. The posture of the whole body at the end time of the i-th time segment is regarded as a key pose in the i-th time segment of choreography, and the crown, neck joint, right shoulder joint, right elbow joint, right wrist joint, right hand tip, left shoulder joint 19 of the left elbow joint, left wrist joint, left hand tip, hip-chest joint, right hip joint, right knee joint, right ankle joint, right toe part, left hip joint, left knee joint, left ankle joint, left toe part For each of the points, the magnitude of the difference between the position in the exemplary motion data DATA _{MDL and} the position in the body motion data DATA _MOT is calculated, and an average value D (i) of the position differences of all 19 points is derived. This value is regarded as indicating the degree of acquisition of the learner's evaluation item key pose in the i-th time segment, and is output as an evaluation value VL _EST having 0 as an ideal value.

With respect to the evaluation value VL _EST , the evaluation value of choreography is calculated as follows by separating the movement of the upper body and the movement of the lower body. Exemplary motion data DATA _MDL for each of the seven points of the crown, right elbow joint, right wrist joint, right hand tip, left elbow joint, left wrist joint and left hand tip at each time in the i-th time segment The magnitude of the difference between the position at and the position at the physical motion data DATA _MOT is calculated, and the average value of the positions at all seven points is derived. The average value C _U (i) of the average values of all times in the i-th time segment is calculated, and this value indicates the degree of acquisition of the evaluation item choreography for the learner's upper body movement in the i-th time segment And an evaluation value VL _EST with 0 as an ideal value is output.

On the other hand, at each time in the i-th time segment, in each of the six points of the right knee joint, the right ankle joint, the right toe portion, the left knee joint, the left ankle joint, and the left toe portion, in the exemplary motion data DATA _MDL The magnitude of the difference between the position and the position in the operation data DATA _MOT is calculated, and the average value of the position differences of all six points is derived. The average value C _L (i) of the average values of all times in the i-th time segment is calculated, and this value indicates the degree of acquisition of the evaluation item choreography for the movement of the learner's lower body in the i-th time segment. As an evaluation value, an evaluation value VL _EST with 0 as an ideal value is output.

Regarding the evaluation value VL _EST , the evaluation value of cooperation is calculated as a difference between appearance times of body postures in the upper body motion and the lower body motion as follows.
The appearance time of the body posture of the upper body of the motion data DATA _MOT with respect to the model motion data DATA _MDL indicated by the upper body posture time difference data DATA _{DIF_UB} output from the time difference calculation device 1312 between the model data / learner data at each time in the i-th time segment And the difference between the appearance time of the body posture of the lower body of the motion data DATA _MOT with respect to the motion data DATA _MDL of the lower body posture time difference data DATA _{DIF_LB} output from the time difference calculation device 1312 between the model data and the learner data Deriving the magnitude of the difference between
An average value E (i) of all differences in the i-th time segment is calculated, and this value is regarded as indicating the degree of acquisition of evaluation item cooperation in the i-th time segment, and 0 is an ideal value. The evaluation value VL _EST is output.

_{Regarding the} evaluation value VL _EST , the evaluation value of timing is calculated as the difference between the appearance time of the corresponding body posture in the learner's motion data DATA _{MOT with} respect to the appearance time of each body posture in the exemplary motion data DATA _MDL .
Difference of appearance time of body posture of motion data DATA _MOT with respect to model motion data DATA _MDL indicated by whole body posture time difference data DATA _{DIF_TB} output by model data / learner data time difference calculation device 1312 at each time in i-th time segment The size of is derived. An average value T (i) of all differences in the i-th time segment is calculated, and this value is regarded as indicating the degree of acquisition of the evaluation item timing in the i-th time segment, and 0 is an ideal value. Output as evaluation value VL _EST .

For the evaluation value VL _EST , the evaluation value of smoothness is calculated as follows.
At each time in the i-th segment, the crown, right elbow, right wrist, right hand, left elbow, left wrist, left hand, right knee, right ankle, right toe, left For each of the 13 points of the knee joint, the left ankle joint, and the left toe, the magnitude of the difference between the average acceleration of all 13 points of the exemplary motion data DATA _MDL and the average acceleration of all 13 points of the motion data DATA _MOT is derived. . The average value A (i) of all the differences in the i-th time segment is calculated, and this value is regarded as indicating the degree of mastery of the smoothness of the evaluation item in the i-th time segment. The evaluation value VL _EST as an ideal value is output.

FIG. 5 is a configuration diagram in which the evaluation device 132 derives a comprehensive evaluation score.
In FIG. 5, the hierarchical structure neural network 1321 includes an input layer unit group 1322, an intermediate layer unit group 1323, and an output layer unit group 1324.
All the units in the input layer unit group 1322 are combined with all the units in the intermediate layer unit group 1323, each having a specific combination weight value, and all the units in the intermediate layer unit group 1323 are each specified by a specific unit. The output layer unit group 1324 is coupled with a coupling weight value. The evaluation value VL _EST (j, k) output from the item evaluation unit 131 is input to the neural network 1321 of the comprehensive evaluation unit 132, and the input value x _i to the i-th unit in the input layer unit group 1322 is determined. . When the input values to all M units in the input layer unit group 1322 are determined, in the jth unit in the intermediate layer unit group 1323, the i th unit and the j th unit in the intermediate layer unit group 1323 are An intermediate value y _j of the following equation is calculated using the joint weight value w _ij between the two.
y _j = 1 / [1 + exp {− (w _1j x ₁ + w _2j x ₂ +... + w _Mj x _M )}]

When the intermediate value is calculated in all K units in the intermediate layer unit group 1323, the output layer unit group 1324 has a coupling between the jth unit in the intermediate layer unit group 1323 and the output layer unit group 1324 unit. An output value z of the following equation is calculated using the weight value v _j , and the value becomes a total evaluation score TOTAL _EST with a maximum score of 100 points.
z = v ₁ y ₁ + v ₂ y ₂ +... + v _K y _K

Learning work is performed in advance on the hierarchical neural network 1321 as follows.
Each sample group of motion data DATA _MOT acquired in advance is input to the item evaluation unit 131 to obtain a sample evaluation value group, and at the same time, a dance instructor for each animation created using the sample group Evaluates to a maximum of 100 points to obtain a comprehensive evaluation score sample group. Then, by using the sample evaluation value group as input teacher data and the comprehensive evaluation score sample group as output teacher data, a connection weight value between units in the hierarchical structure neural network 1321 is derived by the back propagation method.

FIG. 6 shows a display screen DSPL by the evaluation result display device 14.
In this embodiment, the evaluation is displayed as an evaluation value VL _EST for each evaluation item on the display screen DSPL, and a radar chart CHRT for each evaluation is displayed. Moreover, the comprehensive score is described in the “score” column as a comprehensive evaluation.
Below the display screen DSPL, an indicator IDC indicating the number and number of beats and a slider SLD for moving the display to a desired number of desired beats are provided.
In FIG. 6, “size of movement”, “key pose”, “choreography”, “coordination of movement”, “timing”, and “smoothness” are listed as evaluation items.
In FIG. 6, the skeletal structure human body model MDL _SKLTN is shown by animation together with the radar chart CHRT and the evaluation value VL _EST . Thereby, the learner can receive the total evaluation of the evaluation items and the comprehensive evaluation of all the evaluation items with respect to the entire series of body movements, and thus can know the defects and achievement points.

DESCRIPTION OF SYMBOLS 1 Body motion learning support apparatus 11 Motion capture apparatus 12 Motion data storage apparatus 13 Evaluation apparatus 14 Evaluation result display apparatus 111 Magnetic sensor group 112 Computer 131 Item evaluation part 132 Comprehensive evaluation part 1311 Time segmentation division apparatus 1312 Model data / learner data Inter-time difference calculation device 1313 Item evaluation device 1321 Hierarchical structure neural network 1322 Input layer unit group 1323 Intermediate layer unit group 1324 Output layer unit group A Learner CHRT Radar chart DATA _{DIF_TB} Whole body posture time difference data DATA _{DIF_UB} Upper body posture time difference data DATA _{DIF_LB} Lower body posture Time difference data DATA _DIV choreographic division time data DATA _MOT operation data DATA _MDL model operation data DATA _PART time segment data ITM _EST (j, k) Evaluation item I DC indicator MDL _SKLTN human body model (skeleton)
SLD slider TOTAL _EST overall evaluation score VL _EST (j, k), VL _EST evaluation value (score or rank)

Claims (4)

In the body movement learning support device that evaluates the body movement of one or more learners acquired by practice and expresses the body movement by animation so that the difference from the model movement can be understood,
A motion capture device for acquiring the physical motion as physical motion data;
A motion data storage device that stores physical motion data acquired by the motion capture device and stores exemplary motion data used for comparison with the physical motion data;
An evaluation device that evaluates the physical motion using a score or rank by comparing the physical motion data stored in the motion data storage device and the exemplary motion data;
An evaluation result display device for expressing the physical motion as the animation using the physical motion data and the exemplary motion data stored in the motion data storage device, and displaying the result of the evaluation by the evaluation device; ,
With
The evaluation apparatus evaluates at least one evaluation item for each part divided on the time axis for the series of body movements, or further, at least one evaluation item for the whole series of body movements Evaluate
A body movement learning support device characterized by the above.   The body motion learning support apparatus according to claim 1, wherein the combination of the evaluation items is different for each of the parts divided on the time axis.   The physical motion learning support device according to claim 1, wherein the part and / or the evaluation item divided on the time axis are determined by the learner.   The body motion learning support apparatus according to claim 1, wherein the body motion and the model motion are dances.