Coordinate conversion method, coordinate conversion device, motion analysis method, motion analysis device, and computer program

Abstract

The purpose of the present invention is to provide a coordinate conversion method capable of analyzing actions such as walking in detail using action information collected from a single position. This coordinate conversion method is characterized by performing the following steps: acquiring three-dimensional skeleton information of a subject being operated, said three-dimensional skeleton information being described in a first coordinate system comprising an x'direction, a y 'direction, and a z' direction that are orthogonal to each other, and converting the three-dimensional skeleton information into three-dimensional skeleton information using the three-dimensional skeleton information; a Y-direction, which is the direction of travel of the subject, a Z-direction along the torso of the subject, and an X-direction orthogonal to the Y-direction and the Z-direction are calculated, and the first coordinate system is converted into a second coordinate system comprising the X-direction, the Y-direction, and the Z-direction.

Description

Technical Field

[0001] The present invention relates to a coordinate transformation method, a coordinate transformation device, a motion analysis method, a motion analysis device, and a computer program that can be applied to motion analysis such as walking. Background Art

[0002] Markerless motion capture systems are expected to see future growth in various fields due to their ability to simplify motion analysis. However, detailed motion analysis using markerless motion capture systems presents challenges in terms of measurement accuracy and coordinate system definition. In particular, for motions involving movement in a specific direction, such as walking, a coordinate system suitable for analyzing the motion (e.g., a world coordinate system fixed to the ground) is required.

[0003] However, in most markerless motion capture systems, the accuracy of the measured data may deteriorate due to the use of a coordinate system fixed to the system, making it impossible to directly use the measured data for walking motion analysis.

[0004] Alternatively, for example, patent document 1 (Japanese Patent Gazette No. 6381918) discloses an action information processing device, comprising: an acquisition unit, which acquires action information of an object person collected from multiple different positions for each object person performing a prescribed action; a calculation unit, which calculates associated information for locating (position matching) each action information based on the position of the feature point of the object person respectively contained in each action information acquired by the acquisition unit; and an output control unit, which controls the output unit to output the output information after the each action information is located based on the associated information calculated by the calculation unit.

[0005] However, since this device requires motion information collected from multiple locations, complex processing is required to locate each piece of acquired motion information.

[0006] Prior art literature

[0007] Patent Literature

[0008] Patent Document 1: Japanese Patent No. 6381918 Summary of the Invention

[0009] Technical problem to be solved by the invention

[0010] An object of the present invention is to provide a method capable of analyzing in detail the coordinate transformation of an action such as walking using action information collected from one location.

[0011] Technical means to solve the problem

[0012] In order to solve the above-mentioned problem, a first embodiment of the present invention provides a coordinate transformation method, characterized in that:

[0013] Perform the following steps to obtain the skeleton information after coordinate transformation:

[0014] The steps include:

[0015] In a first coordinate system consisting of mutually orthogonal x′, y′, and z′ directions, three-dimensional skeleton information of the subject in motion is acquired.

[0016] Using the skeleton information, calculating a Y direction which is the moving direction of the subject, a Z direction along the torso of the subject, and an X direction orthogonal to the Y direction and the Z direction,

[0017] The first coordinate system is transformed into a second coordinate system consisting of the X direction, the Y direction, and the Z direction.

[0018] In the coordinate transformation method of the present invention, preferably, the Z direction is a direction from the pelvis to the chest of the subject.

[0019] Furthermore, in the coordinate transformation method of the present invention, it is preferable that the transformation into the second coordinate system is performed using a matrix corresponding to the first coordinate system and a matrix corresponding to the second coordinate system.

[0020] A second embodiment of the present invention provides a computer program for executing the following steps to obtain the skeleton information after coordinate transformation:

[0021] The steps include:

[0022] In a first coordinate system consisting of mutually orthogonal x′, y′, and z′ directions, three-dimensional skeleton information of the subject in motion is acquired.

[0023] Using the skeleton information, calculating a Y direction which is the moving direction of the subject, a Z direction along the torso of the subject, and an X direction orthogonal to the Y direction and the Z direction,

[0024] The first coordinate system is transformed into a second coordinate system consisting of the X direction, the Y direction, and the Z direction.

[0025] A third aspect of the present invention provides a coordinate transformation device, comprising:

[0026] an acquisition unit that acquires three-dimensional skeleton information of a subject in motion in a first coordinate system consisting of an x′ direction, a y′ direction, and a z′ direction that are orthogonal to each other;

[0027] an axis direction calculation unit that calculates a Y direction that is a moving direction of the subject, a Z direction along the subject's trunk, and an X direction that is orthogonal to the Y direction and the Z direction, using the skeleton information; and

[0028] The coordinate conversion unit converts the first coordinate system into a second coordinate system consisting of the X direction, the Y direction, and the Z direction.

[0029] A fourth aspect of the present invention provides a motion analysis method, characterized by performing the following steps:

[0030] In a first coordinate system consisting of mutually orthogonal x′, y′, and z′ directions, three-dimensional skeleton information of the subject in motion is acquired.

[0031] Using the skeleton information, calculating a Y direction which is the moving direction of the subject, a Z direction along the torso of the subject, and an X direction orthogonal to the Y direction and the Z direction,

[0032] transforming the first coordinate system into a second coordinate system consisting of the X direction, the Y direction, and the Z direction,

[0033] Perform motion analysis using the transformed skeleton information.

[0034] A fifth aspect of the present invention provides a computer program for executing the following steps:

[0035] In a first coordinate system consisting of mutually orthogonal x′, y′, and z′ directions, three-dimensional skeleton information of the subject in motion is acquired.

[0036] Using the skeleton information, calculating a Y direction which is the moving direction of the subject, a Z direction along the torso of the subject, and an X direction orthogonal to the Y direction and the Z direction,

[0037] transforming the first coordinate system into a second coordinate system consisting of the X direction, the Y direction, and the Z direction,

[0038] Perform motion analysis using the transformed skeleton information.

[0039] A sixth aspect of the present invention provides a motion analysis device, comprising:

[0040] an acquisition unit that acquires three-dimensional skeleton information of a subject in motion in a first coordinate system consisting of an x′ direction, a y′ direction, and a z′ direction that are orthogonal to each other;

[0041] an axis direction calculation unit that calculates a Y direction that is a moving direction of the subject, a Z direction along the subject's torso, and an X direction that is orthogonal to the Y direction and the Z direction, using the skeleton information;

[0042] a coordinate transformation unit that transforms the first coordinate system into a second coordinate system consisting of the X direction, the Y direction, and the Z direction; and

[0043] The analysis unit performs motion analysis using the skeleton information after coordinate transformation.

[0044] Here, each of the above methods is a computer-implemented method. In addition, each of the above computer programs can be stored on a non-transitory computer-readable storage medium and executed by a processor.

[0045] Effects of the Invention

[0046] According to the present invention, since motion information is described in a coordinate system after appropriate conversion, it is possible to analyze motion such as walking in detail using motion information collected from one position. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 1 is a schematic diagram showing the overall configuration of a system including the motion analysis device 10 according to this embodiment.

[0048] Figure 2 It is a block diagram showing the software configuration of the motion analysis device 10 .

[0049] Figure 3 1 is a block diagram showing the hardware configuration of the computer 100 constituting the motion analysis device 10 .

[0050] Figure 4 This is a flowchart showing the steps of motion analysis.

[0051] Figure 5 This diagram shows an example of joints and their connection relationships in body tracking.

[0052] Figure 6 This is a diagram showing an example of a method of defining a coordinate system used in coordinate transformation.

[0053] Figure 7 This figure shows the positions and trajectories of the right foot joints during walking, measured using Azure Kinect and VICON.

[0054] Figure 8A This diagram shows the positions of the center of the shoulder joint and the center of the hip joint measured by Azure Kinect and VICON.

[0055] Figure 8BThis figure shows the trajectory of the center of the shoulder joint during walking, measured using Azure Kinect and VICON.

[0056] Figure 8C This figure shows the trajectory of the center of the hip joint during walking, measured by Azure Kinect and VICON. DETAILED DESCRIPTION

[0057] Representative embodiments of the coordinate transformation method, coordinate transformation device, motion analysis method, motion analysis device, and computer program of the present invention are described in detail below with reference to the accompanying drawings. However, the present invention is not limited to these drawings. Furthermore, the drawings are intended to conceptually illustrate the present invention, and therefore, dimensions, proportions, and numbers may be exaggerated or simplified as necessary to facilitate understanding.

[0058] 1. Regarding the Motion Analysis Device 10 and the Coordinate Transformation Device

[0059] The motion analysis device 10 is a device or system that performs motion analysis using three-dimensional coordinate data obtained by imaging a subject P. Examples of subject P include, but are not limited to, patients undergoing rehabilitation, patients undergoing examinations, and athletes. Specifically, subject P is in motion, for example, by walking or running, or performing some type of motion such as jumping or gymnastics.

[0060] like Figure 2 As shown, the motion analysis device 10 includes an acquisition unit 11, an axis direction calculation unit 13, a coordinate conversion unit 15, and an analysis unit 17. Of these components, the acquisition unit 11, the axis direction calculation unit 13, and the coordinate conversion unit 15 constitute a coordinate conversion device 20.

[0061] The acquisition unit 11 acquires three-dimensional skeleton information of the moving subject P. Here, the skeleton information includes three-dimensional coordinate data.

[0062] The acquisition unit 11 may acquire (ie, generate) the skeleton information originally, or may acquire the skeleton information secondarily by forwarding or the like.

[0063] In the former case, the acquisition unit 11 can be configured as a body tracking system, which can use an imaging device such as an IR camera (depth sensor) to acquire image data containing depth information, and generate a human skeleton model (skeleton) based on the image data (see Figure 5 ), measure the three-dimensional coordinates of the joint points.

[0064] Alternatively, in the latter case, the acquisition unit 11 may be a communication interface for receiving three-dimensional skeleton information from such a body tracking system, or a memory for storing the skeleton information.

[0065] As an example of a body tracking system, Azure Kinect (trademark) of Microsoft Corporation can be cited, but it is not limited to this. For example, it can also be Intel RealSense Depth (trademark) camera of Intel Corporation, ASTRA (trademark) 3D camera of Orbbec 3D Tech.Intl.Inc., Xtion (trademark) of ASUSTeK Computer Inc., and ZED (trademark) camera of Stereolabs Inc.

[0066] The skeleton information acquired by the acquisition unit 11 is described in a coordinate system fixed to the body tracking system (hereinafter sometimes referred to as the original coordinate system or the first coordinate system). For example, in Azure Kinect, the origin of the three-dimensional coordinate value is set to the center of the IR camera (depth sensor), and the coordinate axes are set to the horizontal axis as the x' axis, the vertical axis as the y' axis, and the depth axis as the z' axis when viewed from the front of the IR camera (see Figure 1 ).

[0067] The acquisition unit 11 acquires the skeleton information at a predetermined time interval. The predetermined time interval may be the same as when the body tracking system generates image data or skeleton information, for example, 30 fps.

[0068] Next, the axis direction calculation unit 13 uses the acquired skeletal information to calculate the axis directions of a new coordinate system (hereinafter sometimes referred to as the second coordinate system). Specifically, the axis directions of the new coordinate system are the Y direction, which is the direction of movement of the subject P; the Z direction, which is along the torso of the subject P; and the X direction, which is orthogonal to the Y and Z directions. The Z direction is preferably the direction from the pelvis to the chest of the subject P. The designation of the X, Y, and Z directions is arbitrary.

[0069] In this embodiment, the Y direction is set to the direction from the position of the SPINE_NAVAL (the umbilical point, i.e., the lumbar vertebra at the level of the umbilicus) before walking begins to the position of the SPINE_NAVAL after walking ends. In other words, the Y direction can be obtained by subtracting the coordinates of the SPINE_NAVAL before walking begins from the coordinates of the SPINE_NAVAL after walking ends.

[0070] The Z' direction is set to the direction from the PELVIS (pelvis) position to the SPINE_NAVAL position. Specifically, the Z' direction can be obtained by subtracting the PELVIS coordinates from the SPINE_NAVAL coordinates at the same moment. The Z' direction can be calculated by averaging multiple Z' directions calculated within a specified period.

[0071] Furthermore, the X direction is determined to be perpendicular to the Y direction and the Z′ direction, and further, the Z direction is determined to be perpendicular to the X direction and the Y direction.

[0072] Furthermore, the calculated vector indicating the X direction, the vector indicating the Y direction, and the vector indicating the Z direction may also be normalized to a unit length.

[0073] Next, the coordinate conversion unit 15 converts the original coordinate system describing the skeleton information into a new coordinate system having coordinate axes in the X, Y, and Z directions. For example, the coordinate conversion unit 15 can perform the conversion to the new coordinate system using a matrix corresponding to the original coordinate system and a matrix corresponding to the new coordinate system.

[0074] Here, consider the case of transforming the position p of a point in an original coordinate system whose axes are three mutually orthogonal unit vectors i', j', and k' to a new coordinate system whose axes are three other orthogonal unit vectors i, j, and k, sharing the same origin as the original coordinate system. If the coordinates of position p in each coordinate system are (x', y', z') and (x, y, z), position p is expressed as follows.

[0075] [Formula 1]

[0076] p=xi+yj+zk=x′i′+y′j′+z′k′

[0077] This formula can be written using matrices as follows.

[0078] [Formula 2]

[0079]

[0080] Therefore, in the new coordinate system, the position p(x, y, z) is obtained by the following formula.

[0081] [Formula 3]

[0082]

[0083] Here, since the matrices (i, j, k) and (i′, j′, k′) are orthogonal matrices, their inverse matrices are equal to their transposed matrices. Therefore, we have the following equation.

[0084] [Formula 4]

[0085]

[0086] Thus, a transformation matrix M is obtained that transforms the position expression (x′, y′, z′) in the original coordinate system into the position expression (x, y, z) in the new coordinate system.

[0087] [Formula 5]

[0088]

[0089] Therefore, by applying the transformation matrix M to the expression (x′, y′, z′) in the original coordinate system from the left, the expression (x, y, z) in the new coordinate system can be obtained.

[0090] The analyzing unit 17 performs motion analysis using the coordinate-transformed skeleton information.

[0091] Examples of motion analysis include walking analysis, running analysis, swimming analysis, jumping analysis, and gymnastics analysis. In addition, analysis of the above-mentioned motions (movements such as walking, running, or swimming, or certain motions such as jumping or gymnastics) may be considered, such as analysis suitable for running, standing long jump, and the like.

[0092] Among them, analysis of actions such as walking, running, swimming and jumping includes, for example, walking cycle, walking speed, joint angles, posture, stride, step width, foot height, arm (elbow) swing, arm (elbow) height, left and right balance, front and back / left and right shaking of the body, etc., but is not limited to these.

[0093] The results of the motion analysis are displayed on a monitor, printed, or sent to other computers for use by physicians, physical therapists, researchers, etc. These individuals can consider appropriate treatment methods for individual subjects P based on the results of the motion analysis and effectively utilize them for treatment, rehabilitation, and other purposes.

[0094] 2. About the Computer 100 Constituting the Motion Analysis Device 10 and the Coordinate Transformation Device 20

[0095] like Figure 3 As shown, the computer 100 includes a processor 101, a memory 103, and a communication interface 105, and may further include an input device 107 and an output device 109. The computer 100 may be composed of one computer or multiple computers.

[0096] The processor 101 reads various programs and data into the memory 103 and executes them, thereby realizing various functions of the motion analysis device 10 and the coordinate conversion device 20. The processor 101 can be composed of a semiconductor integrated circuit such as a central processing unit (CPU), a graphics processing unit (GPU), or a microprocessor.

[0097] The memory 103 is a random access memory (RAM) and a read-only memory (ROM) that stores various data and programs. The memory 103 includes a non-transitory computer-readable storage medium such as a hard disk drive, a solid-state drive, or a flash memory.

[0098] The communication interface 105 is an interface for connecting to wired and wireless communication networks, such as an adapter for connecting to Ethernet (registered trademark), a modem for connecting to a public telephone line network, a wireless communication device for wireless communication, a USB (Universal Serial Bus) connector for serial communication, an RS232C connector, etc.

[0099] The input device 107 is for inputting various data, for example, a keyboard, a mouse, a touch panel, buttons, a microphone, etc. The output device 109 is for outputting various data, for example, a display, a printer, a speaker, etc.

[0100] 3. Motion analysis method and coordinate transformation method

[0101] Reference Figure 4 The motion analysis method including the coordinate transformation method is described below. The motion analysis method is a method implemented by a computer. Among the steps constituting the motion analysis method, steps S11 to S13 correspond to the coordinate transformation method.

[0102] In step S11, the skeleton information of the subject P is obtained. At this time, the skeleton information is described using a coordinate system (X′, Y′, Z′ axes; unit vectors i′, j′, k′) fixed to the body tracking system or camera (see Figure 1 ). In addition, the skeleton information is acquired at a predetermined time interval.

[0103] Next, in step S12, the X, Y, and Z axis directions (unit vectors i, j, k) of the new coordinate system are calculated based on the skeletal information. For example, the Y direction is the moving direction, the Z direction is the direction along the torso of the subject P, and the X direction is the direction orthogonal to the Y and Z directions.

[0104] Then, in step S13, the coordinate transformation of the skeleton information is performed. For example, the transformation matrix M is generated and applied to the original coordinate system expression (x', y', z') from the left to obtain the new coordinate system expression (x, y, z).

[0105] Furthermore, in step S14 , motion analysis such as walking analysis is performed using the transformed coordinates.

[0106] Then, a series of processing ends.

[0107] 4. Effects of this embodiment

[0108] As can be seen from this embodiment, the inventors defined a new coordinate system for motion analysis based on motion data such as walking measured by a motion capture system such as Azure Kinect, and developed an algorithm for coordinate transformation.

[0109] In analyzing walking motion using existing markerless motion capture systems, these issues are addressed by focusing on parameters such as joint angles and angular velocities that are independent of a coordinate system, or by pre-acquiring information for coordinate transformation (e.g., using markers installed in the laboratory). This embodiment, by acquiring the coordinate transformation information necessary for motion analysis from measurement data, enables detailed analysis of movements such as walking without the need for complex systems.

[0110] 5.Analysis example

[0111] Here, walking analysis is taken as an example of motion analysis, and the results of verifying the accuracy of the motion analysis according to the present embodiment are shown.

[0112] To evaluate the accuracy of the walking analysis, comparison was made with measurement data from the VICON (trademark) system manufactured by Vicon Motion Systems Ltd. The VICON is a motion capture system that uses an infrared camera to track retroreflective markers attached to the body of the subject P and is considered to provide accurate and highly reliable data.

[0113] In this analysis, Azure Kinect (trademark) is used as the motion capture system. Azure Kinect is a markerless motion capture system that projects an infrared dot pattern onto the subject to obtain a depth image, automatically fits a human skeleton model (skeleton) from the depth image, and measures the coordinates of 20 joint points (see Figure 5 ).

[0114] In Azure Kinect, the built-in processor estimates the joints of the body and reads the three-dimensional coordinate values ​​(x, y, z) of the joint data when constructing the skeleton information. The origin of the three-dimensional coordinate values ​​read from the joint data is the center of the IR camera (depth sensor). The coordinate axis is the horizontal axis X, the vertical axis Y, and the depth axis Z when viewed from the front of the IR camera (refer to Figure 1 ). In addition, IR cameras use infrared light to measure the distance (depth) from the camera to an object.

[0115] In Azure Kinect, the coordinate origin is at the camera's focal point. This coordinate system has a positive X-axis pointing right, a positive Y-axis pointing downward, and a positive Z-axis pointing forward. Therefore, in Azure Kinect, joint positions are represented by the depth sensor's relative values ​​relative to a reference frame. In other words, the skeletal information of subject P is represented by a coordinate system fixed to Azure Kinect.

[0116] Here, the subject P can move in a desired direction relative to the camera of the motion capture system. In this case, in the above coordinate system, the measurement accuracy of the joints that are blocked relative to the camera may deteriorate depending on the movement direction of the subject P.

[0117] In addition, the axes of the new coordinate system are set as follows (see Figure 6 ).

[0118] Y direction: The direction from SPINE_NAVEL (umbilical point) at the beginning of walking to SPINE_NAVEL (umbilical point) at the end of walking

[0119] Z′ direction: the direction from PELVIS (pelvic point) to SPINE_CHEST (thoracic spine point) during walking (average)

[0120] X direction: The direction perpendicular to the Y axis and the Z' axis

[0121] Z direction: The direction perpendicular to the X and Y axes

[0122] 5-1. Measurement of foot joint position during walking

[0123] The positions of the foot joints of the walking subject P were measured using Azure Kinect and VICON, and their trajectories were compared. The Azure Kinect measurement results were subjected to the aforementioned coordinate transformation.

[0124] An example of the measurement results for the right foot joint is shown in Figure 7 The example shown in the figure shows the measurement results when the subject P walks diagonally forward relative to the camera, and the original coordinate system (x′, y′, z′ axes) is shown as a reference.

[0125] Table 1 shows the correlation coefficients of the positions of the left and right foot joints during walking measured by Azure Kinect and VICON.

[0126] [Table 1]

[0127]

[0128] Generally, a positive correlation is considered to exist when the correlation coefficient r satisfies 0.4 ≤ r ≤ 0.7, and a strong positive correlation is considered to exist when it satisfies 0.7 ≤ r ≤ 1. Therefore, according to Table 1, a strong correlation is considered to exist particularly in the front-back direction and the vertical direction.

[0129] Table 2 shows the root mean square error of the positions of the left and right foot joints during walking measured by Azure Kinect and VICON.

[0130] [Table 2]

[0131]

[0132] As can be seen from Table 2, the errors in the left-right direction and the vertical direction are particularly small.

[0133] In addition, the stride length (average) can be calculated from the trajectory of the foot joint position. Table 3 below shows an example of the left and right stride lengths during walking measured by Azure Kinect and VICON.

[0134] [Table 3]

[0135]

[0136] According to Table 3, the difference between the measured values ​​of Azure Kinect and VICON is in the range of several percentages.

[0137] 5-2. Midpoint of left and right hip joints

[0138] The positions of the midpoints of the left and right shoulder joints and the left and right hip joints of the subject P were measured using Azure Kinect and VICON (refer to Figure 8A ), and compared their trajectories (refer to Figure 8B and Figure 8C ). The measurement results of Azure Kinect implement the above coordinate transformation.

[0139] The root mean square errors of the measured left and right shoulder and hip joint centers are shown in Table 4 below.

[0140] [Table 4]

[0141]

[0142] In addition, the correlation coefficients of the measured shoulder joint centers and hip joint centers are shown in Table 5.

[0143] [Table 5]

[0144]

[0145] According to Table 5, strong correlation is considered to exist particularly in the left-right direction and the front-back direction.

[0146] These comparative examples confirm that the measurement data from Azure Kinect after coordinate transformation has an accuracy comparable to that of VICON. Therefore, it is believed that the measurement results from Azure Kinect after coordinate transformation can accurately represent the joint positions of the subject P.

[0147] Therefore, the coordinate transformation technology disclosed herein can be applied to motion analysis represented by walking analysis.

[0148] Therefore, in order to achieve reliable motion analysis, there is no need to set up multiple cameras or place markers on the ground, etc., so the motion capture system can be easily constructed by using coordinate transformation technology.

[0149] While representative embodiments of the present invention have been described above, the present invention is not limited thereto, and various design modifications are possible, which are also encompassed by the present invention.

[0150] Description of Reference Numerals

[0151] 10Motion analysis device

[0152] 11 Acquisition Department

[0153] 13. Axis direction calculation unit

[0154] 15Coordinate transformation unit

[0155] 17 Analysis Department

[0156] 20 coordinate transformation device.

Claims

1. A coordinate transformation method, characterized in that: Perform the following steps to obtain the skeleton information after coordinate transformation: The steps include: In a first coordinate system consisting of mutually orthogonal x′, y′, and z′ directions, three-dimensional skeleton information of the subject in motion is acquired. Using the skeleton information, calculating a Y direction which is the moving direction of the subject, a Z direction along the torso of the subject, and an X direction orthogonal to the Y direction and the Z direction, The first coordinate system is transformed into a second coordinate system consisting of the X direction, the Y direction, and the Z direction.

2. The coordinate transformation method according to claim 1, wherein: The Z direction is a direction from the pelvis to the chest of the subject.

3. The coordinate transformation method according to claim 1, wherein: The first coordinate system is transformed into the second coordinate system using a matrix corresponding to the first coordinate system and a matrix corresponding to the second coordinate system.

4. A computer program, characterized in that The method is used to perform the following steps to obtain the skeleton information after coordinate transformation, the steps comprising: In a first coordinate system consisting of mutually orthogonal x′, y′, and z′ directions, three-dimensional skeleton information of the subject in motion is acquired. Using the skeleton information, calculating a Y direction which is the moving direction of the subject, a Z direction along the torso of the subject, and an X direction orthogonal to the Y direction and the Z direction, The first coordinate system is transformed into a second coordinate system consisting of the X direction, the Y direction, and the Z direction.

5. A coordinate transformation device, characterized in that: include: an acquisition unit that acquires three-dimensional skeleton information of a subject in motion in a first coordinate system consisting of an x′ direction, a y′ direction, and a z′ direction that are orthogonal to each other; an axis direction calculation unit that calculates a Y direction that is a moving direction of the subject, a Z direction along the subject's torso, and an X direction that is orthogonal to the Y direction and the Z direction, using the skeleton information; and The coordinate conversion unit converts the first coordinate system into a second coordinate system consisting of the X direction, the Y direction, and the Z direction.

6. A motion analysis method, characterized in that: Perform the following steps: In a first coordinate system consisting of mutually orthogonal x′, y′, and z′ directions, three-dimensional skeleton information of the subject in motion is acquired. Using the skeleton information, calculating a Y direction which is the moving direction of the subject, a Z direction along the torso of the subject, and an X direction orthogonal to the Y direction and the Z direction, transforming the first coordinate system into a second coordinate system consisting of the X direction, the Y direction, and the Z direction, Use the transformed skeleton information to perform motion analysis.

7. A computer program, characterized in that Used to perform the following steps: In a first coordinate system consisting of mutually orthogonal x′, y′, and z′ directions, three-dimensional skeleton information of the subject in motion is acquired. Using the skeleton information, calculating a Y direction which is the moving direction of the subject, a Z direction along the torso of the subject, and an X direction orthogonal to the Y direction and the Z direction, transforming the first coordinate system into a second coordinate system consisting of the X direction, the Y direction, and the Z direction, Use the transformed skeleton information to perform motion analysis.

8. A motion analysis device, characterized in that: include: an acquisition unit that acquires three-dimensional skeleton information of a subject in motion in a first coordinate system consisting of an x′ direction, a y′ direction, and a z′ direction that are orthogonal to each other; an axis direction calculation unit that calculates a Y direction that is a moving direction of the subject, a Z direction along the subject's torso, and an X direction that is orthogonal to the Y direction and the Z direction, using the skeleton information; a coordinate transformation unit that transforms the first coordinate system into a second coordinate system consisting of the X direction, the Y direction, and the Z direction; and The analysis unit performs motion analysis using the skeleton information after coordinate transformation.

Images