JP2024066646A - Motion evaluation device, method and program - Google Patents

Abstract

To provide a motion evaluation device that can quantitatively evaluate the second motion by comparing the first motion of a first player with the second motion of a second player.SOLUTION: There is provided a device to evaluate the second motion which is the motion of a second player on the basis of the first motion which is the motion of a first player. The device comprises: an acquisition part that acquires a second motion value obtained by measuring the second motion; an extraction part that extracts the second cooperative motion data showing the characteristic behavior of the second motion by performing singular value decomposition of the second motion value; a first cooperative motion data storage part which stores the first cooperative motion data showing the characteristic behavior of the first motion; a comparison part that obtains the first comparison result 31 obtained by comparing the first cooperative operation data with the second cooperative operation data; and an output part that outputs the first comparison result 31 to a display device 12.SELECTED DRAWING: Figure 6

Description

Application for application of Article 30, Paragraph 2 of the Patent Act has been filed. Publication Name: Symposium: Sports Engineering and Human Dynamics 2022, Proceedings of the Conference, Vol. 31, No. 22 (2022), B-7-2 Publisher: The Japan Society of Mechanical Engineers Publication Date: October 28, 2022 Publication Name: Symposium: Sports Engineering and Human Dynamics 2022, Proceedings of the Conference, Vol. 31, No. 22 (2022), B-7-4 Publisher: The Japan Society of Mechanical Engineers Publication Date: October 28, 2022

The present invention relates to a motion evaluation device, method, and program.

Various motion analysis devices have been proposed for analyzing motions such as golf swings (see, for example, Patent Document 1).

Patent No. 6554158

In recent years, with the rise in health consciousness, various exercises are widely practiced. In a typical exercise lesson, the instructor demonstrates model movements and the student imitates them. However, the instructor's assessment of the student's movements is based solely on the instructor's personal sense, and is not quantitative, making it difficult to grasp areas for improvement in movements.

The present invention was devised in consideration of the above-mentioned circumstances, and its main objective is to provide a device that can quantitatively evaluate a second action by comparing a first action of a first player with a second action of a second player.

The present invention is a device for evaluating a second action, which is an action of a second player, based on a first action, which is an action of a first player, and includes an acquisition unit that acquires a second action value obtained by measuring the second action, an extraction unit that extracts second cooperative action data that indicates characteristic behavior of the second action by performing singular value decomposition on the second action value, a first cooperative action data storage unit that stores the first cooperative action data that indicates characteristic behavior of the first action, a comparison unit that acquires a first comparison result by comparing the first cooperative action data with the second cooperative action data, and an output unit that outputs the first comparison result to a display device.

The action evaluation device of the present invention makes it possible to quantitatively evaluate the second action by comparing the first action of the first player with the second action of the second player.

1 is a configuration diagram conceptually showing an action evaluation system including an action evaluation device according to an embodiment of the present invention. 1 is a block diagram showing an action evaluation device according to an embodiment of the present invention; 4 is a flowchart showing a processing procedure of the action evaluation method according to the present embodiment. 13A is a diagram illustrating position data of each part of the second player's body, and FIG. 13B is position data viewed from the Y-axis direction (side). 11 is a graph comparing the size of a spatial base of first cooperative motion data with the size of a spatial base of second cooperative motion data. FIG. 13 is a diagram showing an example of a first comparison result displayed on a display device. 13 is a flowchart showing a processing procedure of a motion evaluation method according to another embodiment of the present invention. 13 is a graph comparing the size of a spatial base of first cooperative motion data with the size of a spatial base of third cooperative motion data. FIG. 11 is a diagram showing an example of a second comparison result displayed on the display device.

Below, an embodiment of the present invention will be described with reference to the drawings. It should be understood that the drawings contain exaggerated expressions and expressions that differ from the dimensional ratios of the actual structure in order to aid in understanding the contents of the invention. Furthermore, the same reference numerals are used for identical or common elements throughout each embodiment, and duplicate explanations will be omitted. Furthermore, the specific configurations shown in the embodiments and drawings are intended to aid in understanding the contents of the present invention, and the present invention is not limited to the specific configurations shown in the drawings.

[Motion analysis system]
In the action evaluation system (action evaluation device) of this embodiment, a second action which is an action of a second player is evaluated based on a first action which is an action of a first player. Fig. 1 is a configuration diagram conceptually showing an action evaluation system 1 including an action evaluation device 2 of this embodiment.

The actions (first action and second action) are not particularly limited as long as they are actions performed by a player (human body). Examples of actions include the actions of a player performing exercises (exercise actions) and the actions of a player performing sports (sports actions). Actions in this embodiment include exercise actions. Exercise actions include, for example, actions in gymnastics, dance, yoga, tai chi, and the like. Exercise actions in this embodiment are exemplified by gymnastics actions.

The first player (not shown) is not particularly limited as long as he or she is a player who performs a first action to be compared with a second action. In this embodiment, the first player is, for example, an instructor who gives lessons on exercise actions. In this case, the first action in this embodiment is an exemplary action of the second action performed by the second player P2.

The first player is not limited to an instructor, but may be, for example, a more skilled (advanced) exerciser than the second player P2, or a player competing with the second player P2 in terms of superiority of exercise movements.

The second player P2 is a player whose second movement is evaluated by the movement evaluation system 1 (movement evaluation device 2). In this embodiment, the second player P2 is exemplified as a student of an exercise movement lesson (e.g., a beginner or intermediate player). In this case, the second movement is a movement that imitates the first movement by the first player (instructor). Note that the second player P2 is not limited to a student, and may be, for example, a player competing with the first player to see which exercise movement is superior.

The movement evaluation system 1 of this embodiment includes a movement evaluation device 2 and a measurement device 3. The movement evaluation system 1 (movement evaluation device 2) is used to execute the movement evaluation method described below.

[Measuring equipment]
The measurement device 3 is for measuring the second action by the second player P2. The measurement device 3 in this embodiment is configured as a motion capture system 3A. In this embodiment, the measurement device 3 (motion capture system 3A) is also used to measure the first action by the first player. Note that the measurement device 3 is not limited to the motion capture system 3A as long as it can measure the first action and the second action, and various devices can be adopted.

The motion capture system 3A of this embodiment includes multiple cameras 4. These cameras 4 are positioned so that the second action by the second player P2 can be captured from various directions, making it possible to perform three-dimensional measurement of the second action. For example, a three-dimensional motion analysis system manufactured by VICON Corporation can be suitably used for the motion capture system 3A.

The measuring device 3 is communicatively connected to the movement evaluation device 2 via a wired or wireless communication line (not shown). This allows the measurement values measured by the measuring device 3 to be transmitted to the movement evaluation device 2. Note that the transmission of the measurement values is not limited to such a communication line, and may be transmitted, for example, via a communication network. Communication networks include, for example, a wide area network (WAN) and a local area network (LAN). Furthermore, when the measuring device 3 is not connected to the movement evaluation device 2, the position data may be input to the movement evaluation device 2, for example, via a storage medium (not shown) such as a flash memory.

[Motion evaluation device]
The action evaluating device 2 is configured by, for example, a computer 5. Examples of the computer 5 include a desktop computer, a notebook computer, a tablet computer, a smartphone, and a cloud server. A desktop computer is adopted as the computer 5 in this embodiment. Fig. 2 is a block diagram showing the action evaluating device 2 of this embodiment.

The action evaluation device 2 of this embodiment is configured to include, for example, an input device 11, a display device 12, a communication device 13, and a processing device 14.

[Input device/display device]
The input device 11 may be, for example, a keyboard 11a or a mouse 11b shown in Fig. 1. The display device 12 may be, for example, a display 12a shown in Fig. 1.

[Communication device]
2, the communication device 13 of this embodiment is communicatively connected to the measurement device 3 (motion capture system 3A) via a communication line 23. This allows the communication device 13 (movement evaluation device 2) to acquire (receive) the measurement values of the measurement device 3 via the communication line 23. Furthermore, the communication device 13 (movement evaluation device 2) of this embodiment can transmit, for example, a signal for controlling the measurement device 3 to the measurement device 3 via the communication line.

[Processing device]
The arithmetic processing device 14 of this embodiment is configured to include, for example, a calculation unit (CPU) 15 that performs various calculations, a storage unit 16 in which data, programs, etc. are stored, and a working memory 17.

[Memory unit]
The storage unit 16 is a non-volatile information storage device formed of, for example, a magnetic disk, an optical disk, an SSD, etc. The storage unit 16 in this embodiment includes a data section 18 and a program section 19.

[Data section]
The data section 18 is for storing data (information) necessary for evaluating the second action by the second player P2 shown in Fig. 1, the evaluation result, and the like. The data section 18 of this embodiment includes a first action value storage section 18a, a second action value input section 18b, a cooperative action data input section 18c, and a comparison result input section 18d. Note that the data section 18 is not limited to such an embodiment, and some of these may be omitted, or a data section in which other data is stored may be included. Details of the data input to these data sections 18 will be described later.

[Program section]
The program unit 19 is a program (computer program) necessary for evaluating the second action by the second player P2 shown in Fig. 1. The program unit (program) 19, when executed by the calculation unit 15, can cause the computer 5 to function as a specific means.

The program unit 19 in this embodiment includes an acquisition unit 19a, an extraction unit 19b, a comparison unit 19c, an output unit 19d, and an evaluation unit 19e. Note that the program unit 19 is not limited to this embodiment, and may include program units having other functions. The functions of these program units 19 will be described in detail later.

[Movement evaluation method (first embodiment)]
Next, the action evaluation method of this embodiment will be described. In the action evaluation method of this embodiment, a second action which is the action of the second player P2 shown in FIG. 1 is evaluated based on a first action which is the action of the first player. FIG. 3 is a flowchart showing the processing procedure of the action evaluation method of this embodiment. In this embodiment, each step of the action evaluation method is executed by the action evaluation device 2 (computer 5) shown in FIGS. 1 and 2.

[Get second action value of second player]
In the motion evaluation method of this embodiment, first, a second motion value is obtained by measuring the second motion of the second player P2 shown in Fig. 1 (step S1). The second motion value can be appropriately obtained as long as it can quantitatively express the second motion. The second motion value of this embodiment includes position data of each body part of the second player P2.

In step S1 of this embodiment, first, the acquisition unit 19a included in the program unit 19 shown in FIG. 2 is loaded into the working memory 17. The acquisition unit 19a is a program for acquiring the second operating value. When the acquisition unit 19a is executed by the calculation unit 15, the computer 5 can function as a means for acquiring the second operating value.

In step S1 of this embodiment, first, the second movement by the second player P2 shown in FIG. 1 is measured. In this embodiment, the measurement device 3 (motion capture system 3A) is used to measure the second movement. The start and end of the measurement by the measurement device 3 may be controlled by the movement evaluation device 2 (computer 5) or by an operator, etc.

In this embodiment, prior to measuring the second movement, multiple markers (not shown) are attached to predetermined parts of the body of the second player P2 shown in FIG. 1. In this embodiment, the multiple markers are formed, for example, as light-reflective spheres. Such markers allow the behavior of the body of the second player P2 to be captured with high accuracy by the multiple cameras 4.

The markers (not shown) in this embodiment are attached to various parts of the body of the second player P2 (e.g., head, wrists, fingertips, elbows, shoulders, hips, knees, ankles, and toes, etc.) in the same manner as in the above-mentioned Patent Document 1. In this embodiment, 53 markers are attached, but this is not limited to this embodiment, and for example, some markers may be omitted or other markers may be added depending on the body movement caused by the second action, etc. Each marker is assigned a predetermined identification number (e.g., number 1 to number 53).

Next, from the start to the end of the second action by the second player P2 shown in FIG. 1, the second action is continuously photographed by the multiple cameras 4. This results in an image group (video) including multiple images of the second action photographed continuously in chronological order (at multiple times). The second action is photographed at time intervals corresponding to a predetermined sampling frequency. The sampling frequency can be set appropriately, and is set to 500 Hz, for example.

Each image included in the image group (video) is processed, for example, by image processing software included in the measurement device 3 (motion capture system 3A). This allows the second movement value of the second movement in time series to be obtained.

In this embodiment, position data of a plurality of markers (not shown) is acquired as the second action value. The position data specifies the positions of the body parts of the second player P2 to which the plurality of markers are attached (position data of the body parts). The position data of the plurality of markers (53 markers in this example) may be aggregated into position data of some markers (23 markers in this example) by, for example, averaging the joint centers of the body parts 20 specified by the plurality of markers. This reduces the amount of position data, and suppresses an increase in the number of modes of the collaborative action data obtained by singular value decomposition described below. Even without such aggregation, for example, the position data of the markers attached to the body parts that should be noted when evaluating the second action may be limited among the position data of the plurality of markers (53 markers in this example).

Figure 4 is a diagram explaining the position data of each body part 20 of the second player P2. In Figure 4, (a) is the position data as viewed from the X-axis direction (the front of the second player P2), and (b) is the position data as viewed from the Y-axis direction (the side of the second player P2).

The position data of each body part 20 of the second player P2 is acquired as three-dimensional (Cartesian coordinate system) coordinate values. The three-dimensional coordinate values are identified, for example, based on the depth direction (X-axis direction), left/right direction (Y-axis direction), and height direction (Z-axis direction) when viewed from the front of the second player P2 shown in FIG. 4(a). The second motion value is acquired by identifying such position data of each body part 20 of the second player P2 in chronological order.

The coordinate values (position data) of each body part 20 of the second player P2 may be transformed (moved) so that the coordinate values of one body part 20 selected from the multiple body parts 20 match a predetermined reference position 21. This allows a second motion value in which the coordinate values of each body part 20 are aligned based on the reference position 21 to be acquired. Such a second motion value makes it easy to compare, for example, with the motion values of other players with different physiques (in this example, the first motion value of the first player) or with the motion values of other players whose motions are measured at different positions. The reference position 21 is set, for example, to the origin of the X-axis, Y-axis, and Z-axis, but is not particularly limited and can be set to any position. In addition, the part 20 to be matched to the reference position 21 is set to the left toe 20a, but is not particularly limited and any body part 20 can be selected.

The coordinate values (position data) of each body part 20 of the second player P2 may be transformed so that, for example, a vector (not shown) pointing from the position of the right toe 20b to the position of the left toe 20a of the second player P2 is parallel to the Y-axis direction (left-right direction). This allows a second movement value with the orientation of the second player P2's body aligned to be obtained, making it easier to compare with, for example, the movement values of other players (first movement value of the first player).

The second movement value is defined, for example, by the following formulas (1) to (3). First, in formula (1), position data (position vector with the origin as the reference point) of the i-th marker (each body part 20) is identified at any time t between the first time and the N-th time. In this embodiment, the first time is the time when the second movement is started. Furthermore, the N-th time is the time when the second movement is ended.

ここで、
ｉ：マーカー（身体各部位）の識別番号
ｔ：時刻

here,
i: Identification number of marker (each body part) t: Time

In the above formula (1), x _i (t), y _i (t), and z _i (t) indicate the X-axis coordinate, Y-axis coordinate, and Z-axis coordinate of the i-th marker (body part 20) at time t. Position data r _i (t) is created from the first time to the N-th time, and a matrix [r _i ] in which these are arranged row by row is defined by the following formula (2).

The matrix [ _ri ] in the above formula (2) is a matrix with N rows and 3 columns, and the three-dimensional coordinates _ri (1), _ri (2), ..., ri(N) defined in the above formula (1) are arranged in the first to Nth rows in chronological order. Then, the matrix [R] in which 23 matrices [ _r1 ], [ _r2 ] ... [ _r23 ] for all markers (each body part 20) are arranged in this order in the column direction (horizontal direction) is defined by the following formula (3). The matrix [R] is a matrix with N rows and 69 (= 3 x 23) columns.

The matrix [R] in the above formula (3) is time-series behavior data that represents the positions of 23 markers (body parts 20) from the first time to the Nth time in a time series. Note that the tth row of the matrix [R] contains the three-dimensional coordinates of the 23 markers (body parts 20) on the body of the second player P2 at the tth time.

The t-th row of this matrix [R] represents the position data of each body part 20 of the second player P2 at the t-th time (the posture of the second player P2). The second motion value can be determined by such a matrix [R].

In step S1 of this embodiment, a human body model (stick picture) M2 is set that includes a plurality of body parts 20 identified by the matrix [R] as joints and bones 24 connecting these joints. These second motion values and the human body model M2 are stored in the second motion value input unit 18b (shown in FIG. 2).

[Extracting second cooperative motion data from second motion value]
Next, in the action evaluation method of this embodiment, second cooperative action data indicating characteristic behavior of the second action is extracted (step S2). In step S2 of this embodiment, the second action value is subjected to singular value decomposition to extract the second cooperative action data.

In step S2 of this embodiment, first, the second operation value input to the second operation value input unit 18b shown in FIG. 2 is read into the working memory 17. Furthermore, the extraction unit 19b included in the program unit 19 is read into the working memory 17. The extraction unit 19b is a program for extracting the second cooperative operation data. When the extraction unit 19b is executed by the calculation unit 15, the computer 5 can function as a means for extracting the second cooperative operation data.

In step S2 of this embodiment, the matrix in the above formula (3) is expanded prior to singular value decomposition. This matrix expansion is intended to bring the reference point of the singular value decomposition closer to the initial posture of the second movement (exercise movement).

To expand the matrix, first, the coordinate values (position vector) r _i (1) of the initial posture of the i-th marker (each body part 20 shown in FIG. 4) are found based on the above formula (1). Next, a matrix ([r _i (1)]) is created in which the first time through the 2Nth time are arranged for each row of the found coordinate values (position vector) r _i (1). At this time, all elements of each row of [r _i (1)] become r _i (1).

Next, the above matrix ([r _i (1)]) is created for markers 1 to 23 (body parts 20), and a matrix ([R(1)]) that arranges them column by column is defined by the following equation (4).

Furthermore, in order to increase the number of data points in the matrix and improve the resolution, a matrix [R _{t ] obtained by inverting the matrix of the above formula (3) in the time series direction is defined by the following formula (5). Then, an observation matrix [R a} ] obtained by combining the matrix [R] of the above formula (3) with the matrix [R(1)] of the above formula (4) and the matrix [R _t ] of the following formula ( ₅ ) is defined by the following formula (6).

Next, in step S2 of this embodiment, the time-series data of the positions of the 23 markers (body parts 20) are defined as the observation matrix [R _a ] in the above formula (6), and the matrix [R ₀ ] indicating the reference initial posture is used to perform singular value decomposition using the following formula (7). Note that the matrix [R ₀ ] is defined as a matrix (6N rows and 69 (=3×23) columns) in which the observation matrix [R _a ] is averaged in the time direction (row direction) and 6N averages are arranged per row.

Based on the above equation (7), the difference between the observation matrix [R _a ] and the initial posture matrix [R ₀ ], [R _a ]-[R ₀ ], is subjected to singular value decomposition to obtain U _(n) , Γ _(n) , and V _(n) for n = 1, 2, ..., J (J is an integer less than or equal to 69).

Γ _(n) is the singular value of the nth mode (nth) and represents the ratio of the nth mode (nth) to [R _a ]-[R ₀ ]. U _(n) is the left singular vector of the observation matrix [R _a ] and has 6N rows and 69 (=3×23) columns. _{V (n)} is the right singular vector of the observation matrix [R _a ] and is 69-dimensional.

The left singular vector U _(n) is a time basis that is a basis representing the time information of the right singular vector V _(n) . On the other hand, the right singular vector V _(n) is a spatial basis that is a basis representing the position information of each marker (each body part 20) on the body of the second player P2. This spatial basis is a vector that indicates the direction of the n-th mode (n-th) movement with respect to the matrix [R ₀ ] of the reference initial posture. The value of such a spatial basis makes it possible to digitize the movement (movement of each body part 20) of the second player P2 in the second movement for each mode.

In step S2 of the present embodiment, a spatial basis (right singular vector V _(n) ) obtained by singular value decomposition of the second action value is extracted as second cooperative action data indicating characteristic behavior of the second action. The second cooperative action data is stored in the cooperative action data input unit 18c shown in FIG. 2 for each mode (in this example, for each of the first to 69th modes).

[Get first comparison result]
Next, in the evaluation method of this embodiment, a first comparison result is obtained by comparing the first cooperative movement data with the second cooperative movement data (step S3).

The first cooperative action data is an index showing characteristic behavior of the first action (in this example, the model action of the second action), which is the action of the first player (not shown). It is preferable that this first cooperative action data is extracted prior to carrying out the evaluation method.

The first cooperative movement data can be extracted in the same procedure as the second emphasizing movement data (steps S1 and S2). To extract this first cooperative movement data, first, a first movement value obtained by measuring the first movement is obtained. This first movement value includes position data of each part of the first player's body, and is stored in advance in the first movement value storage unit 18a shown in FIG. 2.

Next, the first cooperative motion data can be extracted by performing singular value decomposition on the first motion value. The first cooperative motion data includes a spatial basis (right singular vector V _(n) ). The first cooperative motion data of this embodiment is stored in advance in the cooperative motion data input unit 18c shown in FIG. 2 for each mode (in this example, for each of the first to 69th modes).

In step S3 of this embodiment, first, the first cooperative action data and the second cooperative action data inputted to the cooperative action data input unit 18c shown in FIG. 2 are read into the working memory 17. Furthermore, the comparison unit 19c included in the program unit 19 is read into the working memory 17. The comparison unit 19c is a program for obtaining a first comparison result obtained by comparing the first cooperative action data with the second cooperative action data. When the comparison unit 19c is executed by the calculation unit 15, the computer 5 can function as a means for obtaining the first comparison result.

The comparison between the first cooperative operation data and the second cooperative operation data can be performed as appropriate. In this embodiment, it is preferable to compare the first cooperative operation data and the second cooperative operation data for each of the multiple modes (in this example, the first mode to the 69th mode) included in the cooperative operation data.

In this embodiment, the similarity between the spatial base of the first cooperative motion data (right singular vector V _(n) ) and the spatial base of the second cooperative motion data (right singular vector V _(n) ) is obtained for each of a plurality of modes. By obtaining such similarity as the first comparison result, it becomes possible to quantitatively evaluate the degree to which the second motion (movements of each body part 20) of the second player P2 shown in FIG. 4 is similar to the first motion (movements of each body part) of the first player (not shown).

The similarity is obtained as appropriate if it is possible to compare the first cooperative operation data with the second cooperative operation data. In this embodiment, the similarity includes cosine similarity or correlation coefficient. The cosine similarity of the nth mode is obtained by the following formula (8).

ここで、
Ｖ’_ins(n) ：第１協調動作データの第ｎモードにおける各部位の空間基底をノルム化したベクトル（以下、「空間ベクトルノルム」ということがある。）
Ｖ’_1(n) ：第２協調動作データの第ｎモードにおける各部位の空間基底をノルム化したベクトル
|Ｖ’_ins(n)|：第１協調動作データの第ｎモードにおける全部位の空間基底における
ノルム
|Ｖ’_1(n)| ：第２協調動作データの第ｎモードにおける全部位の空間基底における
ノルム
ｎ：モード特定するための変数

here,
_V'ins(n) : A vector obtained by norming the spatial basis of each part in the n-th mode of the first coordination data (hereinafter, may be referred to as "spatial vector norm").
V' _1(n) : A vector obtained by norming the spatial basis of each part in the n-th mode of the second coordinated motion data
| _V'ins(n) |: norm of all parts in the spatial basis in the n-th mode of the first coordinated motion data
|V' _1(n) |: norm of all parts in the spatial basis in the n-th mode of the second coordinated motion data n: variable for identifying the mode

In the above formula (8), _V'ins(n) is a vector obtained by norming the spatial basis of the first cooperative motion data in the nth mode (right singular vector V _(n) ) with the x-, y-, and z-axis components of each body part (i.e., the components _{V'ins_1(n)} to _{V'ins_23(n)} ). _V'1(n) is a vector obtained by norming the spatial basis of the second cooperative motion data in the nth mode (right singular vector V _(n) ) with the x-, y-, and z-axis components of each body part (i.e., the components _{V1_1(n)} to _{V1_23(n)} ).

In the above formula (8), | _V'ins(n) | is the norm of the spatial basis (right singular vector V _(n) ) of all parts and all components of the first cooperative movement data for the nth mode. This norm | _V'ins(n) | is a value obtained by calculating the sum of the squared values of the spatial basis (right singular vector V _(n) ) of the first cooperative movement data and taking the square root of the sum. Such norm | _V'ins(n) | makes it possible to quantify the magnitude of the movement of the first player (movement of each body part 20) in the first movement (not shown).

In the above formula (8), |V'1 _(n) | is the norm of the spatial basis (right singular vector V _(n) ) of all parts and all components of the second cooperative movement data for the nth mode. This norm | _V'1(n) | is a value obtained by calculating the sum of the squared values of the spatial basis (right singular vector V _(n) ) of the second cooperative movement data and taking the square root of the sum. Such norm | _V'1(n) | makes it possible to quantify the magnitude of the movement (movement of each body part 20) of the second player P2 in the second movement shown in FIG. 4.

In step S3 of the present embodiment, the cosine similarity of the above formula (8) is used to calculate the similarity between a vector ( _V'ins(n) ) obtained by norming the spatial base of each part in the nth mode of the first cooperative movement data and a vector (V'1 _{(n)) obtained by norming the spatial base of each part in the nth mode of the second cooperative movement data for each body part 20, for all modes (in this example, the first mode to the 69th mode} ). By obtaining such cosine similarity as the first comparison result, the second movement that simulates the first movement can be evaluated.

The cosine similarity may also be calculated for multiple parts selected from each body part 20 based on the following formula (9):

ここで、
Ｖ’_{ins(n)_k} ：第１協調動作データの第ｎモードでの部位範囲(ｋ)における各部位の空間基底をノルム化したベクトル
Ｖ’_{1(n)_k} ：第２協調動作データの第ｎモードでの部位範囲(ｋ)における各部位の空間基底をノルム化したベクトル
|Ｖ’_{ins(n)_k} |：第１協調動作データの第ｎモードでの部位範囲（ｋ）の全体におけるノルム
|Ｖ’_{1(n)_k} | ：第２協調動作データの第ｎモードでの部位範囲（ｋ）の全体におけるノルム
ｋ：部位を特定するための変数
ｎ：モード特定するための変数

here,
_{V'ins(n)_k} : A vector obtained by norming the spatial basis of each part in the part range (k) in the n-th mode of the first cooperative motion data. V'1 _{(n)_k} : A vector obtained by norming the spatial basis of each part in the part range (k) in the n-th mode of the second cooperative motion data.
| _{V'ins(n)_k} |: norm of the entire part range (k) in the n-th mode of the first coordinated motion data
|V' _{1(n)_k} |: norm of the whole part range (k) in the n-th mode of the second coordinated motion data k: variable for identifying the part n: variable for identifying the mode

In the above formula (9), for multiple parts constituting a specific part range of the body (for example, the lower body), the cosine similarity of the part specified by variable _m1 to the cosine similarity of the part specified by variable _m2 are added together. From such cosine similarity, the similarity in the specific part range is found.

The correlation coefficient C _(n) is obtained by the following equation (10).

ここで、
Ｖ’_ins(n)：第１協調動作データの第ｎモードにおける各部位の空間基底を
ノルム化したベクトル
Ｖ’_1(n) ：第２協調動作データの第ｎモードにおける各部位の空間基底を
ノルム化したベクトル
ｎ：モード特定するための変数

here,
_V'ins(n) : A vector obtained by norming the spatial base of each part in the n-th mode of the first coordinated motion data. V'1 _(n) : A vector obtained by norming the spatial base of each part in the n-th mode of the second coordinated motion data. n: A variable for identifying the mode.

In the above formula (9), for each n-th mode (in this example, modes 1 to 69), a correlation coefficient is calculated between a vector ( _V'ins(n) ) obtained by norming the spatial base of each part of the first cooperative movement data and a vector (V'1(n)) obtained by norming the spatial base of each part of the second cooperative movement data. Then, the correlation coefficients of all these modes may be averaged to calculate an average value of the correlation coefficients, or a weighted average value may be calculated by setting different weights for each mode.

The correlation coefficient may also be calculated for multiple parts selected from each body part 20 based on the following formula (11):

ここで、
Ｖ’_{ins(n)_k} ：第１協調動作データの第ｎモードでの部位範囲(ｋ)における各部位の空間基底をノルム化したベクトル
Ｖ’_{1(n)_k} ：第２協調動作データの第ｎモードでの部位範囲(ｋ)における各部位の空間基底をノルム化したベクトル
|Ｖ’_{ins(n)_k} |：第１協調動作データの第ｎモードでの部位範囲（ｋ）の全体におけるノルム
|Ｖ’_{1(n)_k} | ：第２協調動作データの第ｎモードでの部位範囲（ｋ）の全体におけるノルム
ｋ：部位を特定するための変数
ｎ：モード特定するための変数

here,
_{V'ins(n)_k} : A vector obtained by norming the spatial basis of each part in the part range (k) in the n-th mode of the first cooperative motion data. V'1 _{(n)_k} : A vector obtained by norming the spatial basis of each part in the part range (k) in the n-th mode of the second cooperative motion data.
| _{V'ins(n)_k} |: norm of the entire part range (k) in the n-th mode of the first coordinated motion data
|V' _{1(n)_k} |: norm of the whole part range (k) in the n-th mode of the second coordinated motion data k: variable for identifying the part n: variable for identifying the mode

In the above formula (11), for a plurality of parts constituting a specific part range of the body (e.g., the lower body), the correlation coefficients of the parts specified by the variable _m1 to the correlation coefficients of the parts specified by the variable _m2 are added together. The similarity in the specific part range is calculated using such correlation coefficients. By calculating these correlation coefficients, the strength of the linear correlation between the first movement and the second movement can be calculated, and the second movement simulating the first movement can be evaluated.

In this embodiment, the cosine similarity is obtained as the first comparison result, but instead of the cosine similarity, a correlation coefficient may be obtained, or both may be obtained. In addition, other parameters different from the cosine similarity and the correlation coefficient may be obtained as the first comparison result. In addition, in this embodiment, the space vector norm is used to calculate the cosine similarity and the correlation coefficient, but this is not limited to this aspect. For example, the cosine similarity and the correlation coefficient may be calculated based on the space vector (right singular vector).

In this embodiment, a mode with a relatively low degree of similarity may be selected from among a plurality of modes (in this example, the first mode to the 69th mode). Such a mode may be identified as a mode that the second player P2 should focus on improving. This mode is obtained as the first comparison result.

The first comparison result may include, for example, a graph comparing the spatial vector norm of the first cooperative motion data with the spatial vector norm of the second cooperative motion data for each marker (each body part 20). FIG. 5 shows a graph 25 comparing the spatial vector norm of the first cooperative motion data with the spatial vector norm of the second cooperative motion data.

In Figure 5, the horizontal axis indicates each body part (marker) 20 shown in Figure 4. "(L)" on this horizontal axis indicates the left part (marker), and "(R)" indicates the left part (marker). Meanwhile, the vertical axis indicates the spatial vector norm, which indicates the magnitude of movement of each body part (marker) 20. Also in Figure 5, "Instructor" indicates the magnitude of movement of each body part of the first player (first cooperative movement data), and "Participant" indicates the magnitude of movement of each body part of the second player (second cooperative movement data).

The graph 25 in FIG. 5 makes it possible to compare the magnitude of the movements made by the first player P2 with the magnitude of the movements made by the second player P2 for each body part 20. Note that FIG. 5 shows the graph 25 in the first mode. The first comparison result is stored in the comparison result input unit 18d shown in FIG. 2.

[Show first comparison result]
Next, in the evaluation method of this embodiment, the first comparison result is output to the display device 12 shown in Figures 1 and 2 (step S4). In step S4 of this embodiment, first, the first comparison result input to the comparison result input unit 18d shown in Figure 2 is read into the working memory 17. Furthermore, the output unit 19d included in the program unit 19 is read into the working memory 17. The output unit 19d is a program for outputting the first comparison result to the display device 12. When this output unit 19d is executed by the calculation unit 15, the computer 5 can function as a means for outputting the first comparison result.

Figure 6 is a diagram showing an example of the first comparison result 31 displayed on the display device 12. In step S4 of this embodiment, the similarity 26 between the spatial base of the first cooperative action data and the spatial base of the second cooperative action data is displayed on the display device 12 as the first comparison result 31. In this embodiment, the cosine similarity is displayed as the similarity 26, but it may be a correlation coefficient or another similarity may be displayed. Also, although the similarity 26 in this embodiment shows the cosine similarity of the first mode, this is not particularly limited, and the cosine similarity may be displayed for each of multiple modes.

In this way, in the motion evaluation method (motion evaluation device 2) of this embodiment, second cooperative motion data showing characteristic behavior of the second motion of the second player P2 (shown in FIG. 1) is extracted, and the second cooperative motion data is compared with the first cooperative motion data of the first player. Then, a first comparison result 31 comparing the first motion with the second motion is output to the display device 12. This allows the second motion of the second player (in this example, the student) P2 to be quantitatively evaluated without relying on the sense of the person evaluating the second motion (in this example, the first player who is the instructor). This makes it easy to grasp the points to be improved in the second motion of the second player P2, and for example, it becomes possible to provide appropriate guidance on the second motion to the second player P2, and the second motion can be further improved (the exercise motion can be established).

In step S4 of this embodiment, a graph 25 (shown in FIG. 5) obtained as the first comparison result 31 may be displayed on the display device 12. Details of the graph 25 are as described above. Such a graph 25 makes it possible to compare the magnitude of the movement of the first action with the magnitude of the movement of the second action for each body part of the first player and the second player P2, and to grasp the movement of each body part that should be focused on for improvement.

In this embodiment, it is preferable that the graph 25 can be switched between multiple modes (mode 1 to mode 69) using, for example, a pull-down menu 27 or the like. This makes it possible to compare the magnitude of movement of each part of the body for each of the multiple modes.

In step S4 of this embodiment, a human body model M1 of the first player and a human body model M2 of the second player P2 may be displayed on the display device 12. Such human body models M1 and M2 make it possible to visually compare the first action and the second action.

In the present embodiment, the human body models M1 and M2 represent the first and second movements at four times among the first time to the Nth time, but are not limited to this aspect. For example, the first and second movements at one time may be represented, or the first and second movements at five or more times may be represented.

In this embodiment, as shown in Figs. 4(a) and (b), for the human body models M1 and M2, the coordinate values of one part 20 selected from the multiple body parts 20 (in this example, the left foot 20a) are matched to the reference position 21. This makes it easy to compare the first and second movements, even if there is a difference in physique between the first player and the second player P2, or a difference in the positions at which the first and second movements are measured, as shown in Fig. 6. In this embodiment, it is preferable that the human body models M1 and M2 can be switched for each of the multiple modes (first mode to 69th mode) using, for example, a pull-down menu 28 or the like. This makes it possible to visually compare the first and second movements for each of the multiple modes.

In step S4 of this embodiment, advice data 29 may be displayed on the display device 12. Such advice data 29 allows the second player P1 (shown in FIG. 1) to specifically grasp the areas that need improvement in the second action, enabling appropriate learning.

To create the advice data 29, first, markers (body parts) in which there is a large difference between the size of the spatial base of the first cooperative movement data and the size of the spatial base of the second cooperative movement data are identified in the graph 25 shown in FIG. 5. Then, if the identified parts are concentrated in the lower body, for example, advice data 29 is created that indicates to increase (or decrease) the movement of the lower body. This makes it possible to grasp more specifically the points to improve in the second movement. Note that the creation of the advice data 29 is not limited to this manner.

In step S4 of this embodiment, video data (not shown) of the first movement may further be output to the display device 12. This allows the second player P2 to learn the first movement that is to be improved with emphasis, and enables the second player P2 to improve the second movement. This video data is preferably video data of a mode with a relatively low degree of similarity among a plurality of modes (in this example, the first mode to the 69th mode). This allows the second player P2 to learn the first movement of the mode that is to be improved with emphasis. In this embodiment, a link 30 to the video data is displayed on the display device 12. This allows easy access to the video data.

In step S4 of this embodiment, for example, video data (not shown) may be displayed that partially extracts the movements of each body part in which the difference between the spatial base size of the first cooperative movement data and the spatial base size of the second cooperative movement data is large in the graph 25 shown in FIG. 5. This allows the second player P2 to focus on the first movement that is necessary to improve the second movement, and to concentrate on practicing that movement.

[Movement evaluation method (second embodiment)]
In the above embodiments, a series of processes of the action evaluation method is completed by outputting the first comparison result 31 obtained by comparing the first cooperative action data with the second cooperative action data, but the present invention is not limited to such an embodiment. For example, a second comparison result obtained by comparing the first cooperative action data with third cooperative action data indicating characteristic behavior of the action of the second player P2 (shown in FIG. 1) learned by viewing the first comparison result 31 may be further output. FIG. 7 is a flowchart showing the processing procedure of the action evaluation method of another embodiment of the present invention.

[Evaluate the second action]
In the action evaluation method of this embodiment, prior to outputting the third cooperative action data, the second action by the second player P2 (shown in FIG. 1) is evaluated as to whether it is good or not (step S5). The second action can be evaluated as appropriate. For example, the second action can be evaluated based on whether the first comparison result 31 satisfies a predetermined criterion. The criterion can be set as appropriate according to, for example, the skill level of the second player P2.

In step S5 of this embodiment, first, the first comparison result 31 (such as the similarity 26 shown in FIG. 6) input to the comparison result input unit 18d shown in FIG. 2 is read into the working memory 17. Furthermore, the evaluation unit 19e included in the program unit 19 is read into the working memory 17. The evaluation unit 19e is a program for evaluating whether the second operation is good or not. When the evaluation unit 19e is executed by the calculation unit 15, the computer 5 can function as a means for evaluating the second operation.

In step S5 of this embodiment, if the similarity 26 (cosine similarity in this example) obtained as the first comparison result 31 shown in FIG. 6 is equal to or greater than a predetermined threshold, the second action is evaluated as being good. The threshold can be set appropriately depending on the skill level of the second player P2 (shown in FIG. 2), etc.

In step S5, if the similarity is equal to or greater than the threshold ("Yes" in step S5), the second action by the second player P2 (shown in FIG. 1) has reached the required level of proficiency. This ends the series of processes in the action evaluation method. On the other hand, in step S5, if it is determined that the similarity is less than the threshold ("No" in step S5), the second action by the second player P2 has not reached the required level of proficiency. In this case, the next steps S6 to S9 are performed.

[Extract third cooperative motion data]
Next, in the movement evaluation method of this embodiment, third cooperative movement data showing characteristic behavior of the movement of the second player P2 (shown in FIG. 1) learned by looking at the first comparison result 31 shown in FIG. 6 is extracted (step S6). The third cooperative movement data is extracted based on the same procedure as steps S1 and S2 in the previous embodiments. Therefore, the acquisition unit 19a and the extraction unit 19b included in the program unit 19 shown in FIG. 2 are executed by the calculation unit 15, so that the computer 5 can function as a means for extracting the third cooperative movement data.

To extract the third cooperative motion data, first, a third motion value is obtained by measuring the third motion of the second player P2 (shown in FIG. 1) who learned by viewing the first comparison result 31 shown in FIG. 6. The third motion value is then subjected to singular value decomposition to extract the third cooperative motion data. This third cooperative motion data includes a spatial basis (right singular vector V _(n) ). The third cooperative motion data is stored in the cooperative motion data input unit 18c shown in FIG. 2.

[Get second comparison result]
Next, in the motion evaluation method of this embodiment, a second comparison result is obtained by comparing the first cooperative motion data with the third cooperative motion data (step S7). In step S7 of this embodiment, first, the first cooperative motion data and the third cooperative motion data inputted to the cooperative motion data input unit 18c shown in FIG. 2 are read into the working memory 17. Furthermore, the comparison unit 19c included in the program unit 19 is read into the working memory 17. The comparison unit 19c is a program for obtaining a second comparison result by comparing the first cooperative motion data with the third cooperative motion data. The comparison unit 19c is executed by the calculation unit 15, so that the computer 5 can function as a means for obtaining the second comparison result.

The second comparison result in this embodiment includes the similarity (in this example, cosine similarity) between the spatial base of the first cooperative motion data and the spatial base of the third cooperative motion data. Furthermore, the second comparison result may include, for example, a graph comparing the size of the spatial base of the first cooperative motion data and the size of the spatial base of the third cooperative motion data for each marker (each body part 20). Figure 8 is a graph 34 comparing the size of the spatial base of the first cooperative motion data and the size of the spatial base of the third cooperative motion data.

In FIG. 8, the horizontal axis indicates each body part (marker) 20, similar to graph 25 in FIG. 5. The vertical axis indicates the spatial vector norm, which indicates the magnitude of movement of each body part (marker) 20. Also in FIG. 8, "Instructor" indicates the magnitude of movement of each body part of the first player (first cooperative motion data). "Before" indicates the magnitude of movement of each body part of the second player (second cooperative motion data) before learning. "After" indicates the magnitude of movement of each body part of the second player (third cooperative motion data) after learning.

Graph 34 in FIG. 8 makes it possible to compare the magnitude of the movement of the first action by the first player with the magnitude of the movement of the second player P2 after learning, for each body part 20. Furthermore, graph 34 shows the magnitude of the movement of the second action by the second player P2 before learning, making it possible to compare the change in the movement of the second player P2 before and after learning, for each body part 20. The second comparison result is stored in the comparison result input unit 18d shown in FIG. 2.

[Show second comparison result]
Next, in the action evaluation method of this embodiment, the second comparison result is output to the display device 12 shown in Fig. 1 (step S8). In step S8 of this embodiment, first, the second comparison result input to the comparison result input unit 18d shown in Fig. 2 is read into the working memory 17. Furthermore, the output unit 19d included in the program unit 19 is read into the working memory 17. The output unit 19d is a program for outputting the second comparison result to the display device 12. When the output unit 19d is executed by the calculation unit 15, the computer 5 can function as a means for outputting the second comparison result.

Figure 9 is a diagram showing an example of the second comparison result 32 displayed on the display device 12. In step S8 of this embodiment, the similarity 33 between the spatial base of the first cooperative action data and the spatial base of the third cooperative action data is displayed on the display device 12 as the second comparison result 32. In this embodiment, the cosine similarity is displayed as the similarity 33, but it may be a correlation coefficient or another similarity may be displayed.

In this manner, in the movement evaluation method (movement evaluation device 2) of this embodiment, the second comparison result 32 is displayed on the display device 12, making it possible to quantitatively evaluate the third movement of the second player P2, who learned the third movement by viewing the first comparison result 31 shown in FIG. 6. This makes it easy to grasp the improvement effect of the third movement, and therefore makes it possible, for example, to provide appropriate guidance to the second player P2 on the third movement and further improve the third movement.

The similarity 26 between the spatial base of the first cooperative action data and the spatial base of the second cooperative action data may also be displayed. This allows the similarity 26 before learning (Before) to be compared with the similarity 33 after learning (After), and the improvement effect of the third action can be accurately understood.

In step S8 of this embodiment, the graph 34 obtained as the second comparison result 32 may be displayed on the display device 12. Details of the graph 34 are as described above. Such a graph 34 makes it possible to compare the magnitude of movement of the first to third movements for each body part of the first and second players P2, and the improvement effect of the third movement can be accurately understood.

In this embodiment, it is preferable that the graph 34 be switchable between multiple modes (mode 1 to mode 69) using, for example, a pull-down menu 35 or the like. This makes it possible to compare the magnitude of movement of each body part for each of the multiple modes, and to grasp the effect of improvement.

In step S8 of this embodiment, a human body model M1 of the first player and a human body model M2 of the second player may be displayed on the display device 12. Such human body models M1 and M2 make it possible to visually compare the first and third actions and grasp the improvement effect. It is preferable that the human body models M1 and M2 can be switched between a plurality of modes (first mode to 69th mode) by a pull-down menu 36 or the like.

In step S4 of this embodiment, advice data 29 may be displayed on the display device 12. Details of the advice data 29 are as described above. If the similarity 33 is good in all modes, as in this embodiment, the indication of areas for improvement may be omitted, or a message conveying that the similarity is good may be displayed.

[Evaluate the third movement]
Next, in the action evaluation method of this embodiment, the third action by the second player P2 (shown in FIG. 1) is evaluated as to whether it is good or not (step S9). The third action is evaluated based on the same procedure as step S5 of the previous embodiments. If it is determined that the third action is not good (step S9: No), steps S6 to S9 are performed again.

In this way, in the movement evaluation method (movement evaluation device 2) of this embodiment, repetitive learning (practice) is carried out until the movement by the second player P2 (shown in FIG. 1) reaches the required level of proficiency. This makes it possible for the movement evaluation method to reliably allow the second player P2 to improve (establish) the exercise movement.

The above describes in detail a particularly preferred embodiment of the present invention, but the present invention is not limited to the illustrated embodiment and can be modified and implemented in various ways.

[Additional Notes]
The present invention includes the following aspects.

[Invention 1]
1. A device for evaluating a second action of a second player based on a first action of a first player, the device comprising:
an acquisition unit that acquires a second action value obtained by measuring the second action;
an extracting unit that extracts second cooperative motion data indicating a characteristic behavior of the second motion by performing singular value decomposition on the second motion value;
a first cooperative motion data storage unit configured to store first cooperative motion data indicating a characteristic behavior of the first motion;
a comparison unit that obtains a first comparison result by comparing the first cooperative operation data with the second cooperative operation data;
an output unit that outputs the first comparison result to a display device;
Behavior evaluation device.
[Invention 2]
2. The action evaluation device according to claim 1, wherein the first action is a model action of the second action.
[The present invention 3]
a first motion value obtained by measuring the first motion includes position data of each body part of the first player;
3. The motion evaluation device according to claim 1, wherein the second motion value includes position data of each body part of the second player.
[Invention 4]
the first cooperative motion data includes a spatial basis obtained by performing singular value decomposition on a first motion value obtained by measuring the first motion;
4. The action evaluating device according to any one of claims 1 to 3, wherein the second cooperative action data includes a spatial basis obtained by singular value decomposition of the second action value.
[Invention 5]
5. The action evaluation device according to claim 4, wherein the comparison unit obtains, as the first comparison result, a similarity between a spatial base of the first cooperative action data and a spatial base of the second cooperative action data.
[Invention 6]
The action evaluation device according to claim 5, wherein the similarity includes a cosine similarity or a correlation coefficient.
[Invention 7]
7. The action evaluation device according to claim 5 or 6, wherein the similarity is acquired for each of a plurality of modes included in the first cooperative action data and the second cooperative action data.
[The present invention 8]
the comparison unit selects a mode having a relatively small degree of similarity from among the plurality of modes,
8. The action evaluation device according to claim 7, wherein the output section further outputs video data of the first action in a selected mode.
[The present invention 9]
the comparison unit further obtains a second comparison result by comparing the first cooperative movement data with third cooperative movement data indicating characteristic behavior of the second player learned by viewing the first comparison result;
9. The action evaluation device according to any one of claims 1 to 8, wherein the output section further outputs the second comparison result.
[Invention 10]
The motion evaluation device according to any one of claims 1 to 9, wherein the first motion and the second motion include an exercise motion.
[Invention 11]
1. A method for evaluating a second action of a second player based on a first action of a first player, the method comprising:
A step of acquiring a second motion value obtained by measuring the second motion by a computer;
extracting second cooperative motion data indicating a characteristic behavior of the second motion by performing singular value decomposition on the second motion value;
acquiring a first comparison result obtained by comparing first cooperative motion data indicating a characteristic behavior of the first motion with the second cooperative motion data;
and outputting the first comparison result to a display device.
Operation evaluation method.
[Invention 12]
A computer program for evaluating a second action of a second player based on a first action of a first player, the computer program comprising:
Computer,
a means for acquiring a second motion value obtained by measuring the second motion;
a means for extracting second cooperative motion data indicating a characteristic behavior of the second motion by performing singular value decomposition on the second motion value;
a means for acquiring a first comparison result obtained by comparing first cooperative motion data indicating a characteristic behavior of the first motion with the second cooperative motion data;
functioning as a means for outputting the first comparison result to a display device;
Computer program.

12 Display device 31 First comparison result

Claims (12)

1. A device for evaluating a second action of a second player based on a first action of a first player, the device comprising:
an acquisition unit that acquires a second action value obtained by measuring the second action;
an extracting unit that extracts second cooperative motion data indicating a characteristic behavior of the second motion by performing singular value decomposition on the second motion value;
a first cooperative motion data storage unit configured to store first cooperative motion data indicating a characteristic behavior of the first motion;
a comparison unit that obtains a first comparison result by comparing the first cooperative operation data with the second cooperative operation data;
an output unit that outputs the first comparison result to a display device;
Behavior evaluation device. The action evaluation device according to claim 1, wherein the first action is a model action of the second action. a first motion value obtained by measuring the first motion includes position data of each body part of the first player;
The action evaluation device according to claim 1 , wherein the second action value includes position data of each body part of the second player. the first cooperative motion data includes a spatial basis obtained by performing singular value decomposition on a first motion value obtained by measuring the first motion;
The action evaluating device according to claim 1 , wherein the second coordination action data includes a spatial basis obtained by performing singular value decomposition on the second action value. The motion evaluation device according to claim 4, wherein the comparison unit obtains, as the first comparison result, a similarity between a spatial base of the first cooperative motion data and a spatial base of the second cooperative motion data. The motion evaluation device according to claim 5, wherein the similarity includes a cosine similarity or a correlation coefficient. The action evaluation device according to claim 5, wherein the similarity is obtained for each of a plurality of modes included in the first cooperative action data and the second cooperative action data. the comparison unit selects a mode having a relatively small degree of similarity from among the plurality of modes,
The action evaluation device according to claim 7 , wherein the output section further outputs video data of the first action in a selected mode. the comparison unit further obtains a second comparison result obtained by comparing the first cooperative movement data with third cooperative movement data indicating characteristic behavior of the second player learned by viewing the first comparison result;
The action evaluation device according to claim 1 , wherein the output section further outputs the second comparison result. The motion evaluation device according to claim 1 or 2, wherein the first motion and the second motion include an exercise motion. 1. A method for evaluating a second action of a second player based on a first action of a first player, the method comprising:
A step of acquiring a second motion value obtained by measuring the second motion by a computer;
extracting second cooperative motion data indicating a characteristic behavior of the second motion by performing singular value decomposition on the second motion value;
acquiring a first comparison result obtained by comparing first cooperative motion data indicating a characteristic behavior of the first motion with the second cooperative motion data;
and outputting the first comparison result to a display device.
Operation evaluation method. A computer program for evaluating a second action of a second player based on a first action of a first player, the computer program comprising:
Computer,
a means for acquiring a second motion value obtained by measuring the second motion;
a means for extracting second cooperative motion data indicating a characteristic behavior of the second motion by performing singular value decomposition on the second motion value;
a means for acquiring a first comparison result obtained by comparing first cooperative motion data indicating a characteristic behavior of the first motion with the second cooperative motion data;
functioning as a means for outputting the first comparison result to a display device;
Computer program.

Images