Suggestion of Motion Velocity/Acceleration Curved Surface

Kaoru Mitsuhashi

, Mizuho Nakamura

, Masanobu Chiba

and Hiroshi Hashimoto

Mechanical Maintenace and Safety Engineering Unit, Polytechnic University of Japan, Kodaira, Tokyo, Japan

Master Program of Innovation for Design and Engineering, Advanced Institute of Industrial Technology, Tokyo, Japan

Keywords: Skill Level, Microsoft Kinect, Motion Curved Surface, Velocity/Acceleration Visualization, RGB Color

Gradation, Experts and Beginners.

Abstract: The skill teaching/succession method is not quantitative but qualitative, which is abstract oral or gesture

expression. Quantitative teaching is difficult for teacher/instructor. In previous research, Expert and

beginner perform the sports and entertainment motion, and the character of the motion curved surface is

analysed using Microsoft Kinect (RGBD camera). The character is the maximum curvature and surface

area. However, the usage of characters is uncertain. In this research, we investigate the correlation of

maximum curvature and surface area from motion curved surface in before and after training. Therefore, we

visualize the different correlation of experts and beginners from the characters and the transition of the skill

training.

1 INTRODUCTION

Motion training of sports, art, and engineering is

difficult quantitatively, because it is almost trial-and-

error. Adjustment of force (pressure) or velocity for

motion skill is necessary, but the method of teaching

for skill is almost oral (onomatopoeia or metaphor)

or watching the movie, they are qualitative (Fujino,

Inoue, Kikkawa, etc., 2005). For example, the

qualitative teaching of force (pressure) adjustment

and muscle tone for motion seems to be effective,

because the nuance of complicated motion is able to

be expressed easily and is impressed strongly.

However, the qualitative expression method

(teaching) is vague, and the instructor can’t make the

learner (beginner) to copy the expert motion

perfectly (Taki, Hasegawa, Fukumura, 1996). Then,

the quantitative teaching is also necessary. The

adjusting velocity/force or muscle tone is measured

or analyzed using acceleration sensor or myogenic

potential sensor, but the measurement using the

sensor can only analyze the local motion. The

showing local motion analysis is difficult to

understand the expert motion intuitively or visually.

On the other hand, only using image processing

seems to be difficult to measure or analyze.

In previous research, we suggest the “Motion

Curve Surface”, it is expressed by human joint

positions and times using Microsoft Kinect. We

evaluate the user motion by its curvature. Firstly, we

visualize a physical motion (human joint trajectory)

using the motion curved surface, and extract the

difference between beginners and experts

(Mitsuhashi, Ohyama, and Hashimoto, 2014),

(Suneya, Kutsuna, and Mitsuhashi, etc., 2014).

Therefore, we can evaluate technical skill

quantitatively, and suggest the skill teaching method

for expert instructors. Secondly, we compose the

motion curved surfaces made from the multiple

Kinect view, so as to track the whole joint motion in

more detail, and confirm the validity of skill

succession by watching skeleton motion movie and

curved surface (Mitsuhashi, Ohyama, and

Hashimoto, 2015). Thirdly, we investigate the

correlation of character from the many motion

curved surfaces. The character of motion curved

surface is maximum curvature and surface area. By

means of expressing the diagram of the motion

curved surface character, the different and tendency

of experts and beginners is extracted. After then, we

investigate the transition of training effectiveness for

beginners from correlation diagram (Mitsuhashi,

Ohyama, and Hashimoto, 2016). However, we can

only clarify the human motion trajectory, but are not

clarifying the timing, rhythm, and force (pressure).

They are necessary to achieve of expert motion. In

addition, expert instructors can teach the skill

motion easily, because they can teach skill important

Mitsuhashi, K., Nakamura, M., Chiba, M. and Hashimoto, H.

Suggestion of Motion Velocity/Acceleration Curved Surface.

DOI: 10.5220/0006477304890493

In Proceedings of the 14th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2017) - Volume 2, pages 489-493

ISBN: Not Available

489

point effectively.

In this paper, we suggest the motion velocity

curved surface and the motion acceleration curved

surface, and the validity is investigated. Firstly, we

explain the calculation (definition) of motion

velocity/acceleration curved surface. The calculation

method is used the differential geometry, and the

express is the RGB color gradation using OpenGL

library. After that, the surfaces are investigated using

some motion samples for the validity. The

differences of expert and beginner motion are

visualized using expert opinions, velocity,

acceleration, and timing from the surfaces.

2 CALCULATION OF MOTION

VELOCITY/ACCELERATION

CURVED SURFACE

The previous motion curved surface is expressed by

the joint trajectories using approximation fitting. The

trajectories are arranged by u-direction of joints and

v-direction of time, and they are captured and

calculated using Microsoft Kinect and Kinect SDK.

We analyze the shape, area, and curvature of motion

curved surface. However, the force adjustment,

timing, and rhythm can’t be expressed using only the

area, shape, and curvature of motion curved surface.

Then, the motion velocity/acceleration curved

surface is suggested using the differential geometry.

Kinect can store the joint positions, but also times.

Consequently, the velocity and acceleration can be

expressed using time derivative (the joint positions

are differentiated by time). When the joint position

u,v

and time t

u,v

is given, the velocity S

u,v

expressed by equation (1).

vuvu

1,,







(1)

Here, S

u,v

is the absolute value because the

direction of S

u,v

is already known. The u-direction of

the joints is not differentiated because it is not

changed by time. The acceleration A

u,v

is expressed

by equation (2) like the velocity calculation.

vuvu

1,,







(2)

Here, A

u,v

is a positive or negative value. In

addition, the each acceleration is changed by

integrating joint mass. After that, the

velocity/acceleration is visualized (expressed) on the

motion curved surface using the RGB color

gradation. The RGB color (r, g, b) of velocity S

u,v

expressed by equation (3).

 

   

2554cos(15.0

255,0,0

)255,,0

)(25.0,255,0

)(50.00,255,

)(75.00,,255

0,0,255

minmax

min,

min

minmaxmin

max

















































temp

SStemp

SSSStemp

bgr



(3)

Here, S

max

is the maximum velocity, S

min

is the

minimum velocity. The parameter r, g, b (0<=r, g,

b<=255) is color strength (red, green, blue). S

max

expressed by only red. S

min

is expressed by only

blue. The acceleration is expressed by the same

method. Therefore, the skill level of the adjustment

force, rhythm, and timing can be visualized. The

previous motion curved surface is expressed by the

velocity or acceleration in the next chapter

(Mitsuhashi, Ohyama, and Hashimoto, 2014) ,

(Mitsuhashi, Ohyama, and Hashimoto, 2015).

Our IDE (Integrated Development Environment)

is Visual Studio Express 2015 for Windows

Desktop. Written program is C++, program library is

OpenGL and Kinect SDK. Subject motion expressed

by skeleton animation from Figure 1, and the motion

curved surface (curvature, velocity, acceleration) is

expressed at the same time. The joints are blue ball,

skeleton (bone) is gray bar. The method of

converting joint points into the curved surface is

approximation fitting.

3 EVALUATION OF MOTION

VELOCITY/ACCELERATION

CURVED SURFACE

The sports motion is expressed by the motion

velocity/acceleration curved surface using the

calculation method of the previous chapter, because

the validity of the surface is investigated. The

motion is throwing in darts, crawl in swimming, and

defense in karate.

ICINCO 2017 - 14th International Conference on Informatics in Control, Automation and Robotics

490

(a) Tracking state

(b) Skeleton animation

Figure 1: Motion tracking state.

3.1 Throw Darts Motion

Figure 2 shows the motion velocity/acceleration

curved surface of expert and beginner in darts throw

motion. They are formed by the joint trajectories of

right hand, right elbow, and right shoulder. In

addition, the gradation color bar is the range of

velocity of acceleration, and added in Figure 2 sides.

Table 1 shows the maximum velocity, the minimum

velocity, the maximum acceleration, the minimum

acceleration, and timing of maximum velocity (or

acceleration) in darts throw motion. The expert has

the fastest in darts release from Figure 2. On the

other hand, the beginner has the fastest in start

throwing; his velocity is constant in darts release.

Expert’s velocity is faster than the beginner’s

velocity from Table 1. Acceleration results are also.

In the expert’s opinion, he needs to let a snap for

wrist in darts release; he needs to concentrate in

wrist or hand. To verify these results, we analyze the

numerical results. Figure 3 shows the velocity and

acceleration of right hand, elbow, and shoulder.

Expert’s maximum velocity/acceleration is in release

darts and beginner’s is in throwing start from

Figure.4. Therefore, we can evaluate also from

velocity, acceleration, and timing.

3.2 Crawl Motion in Swimming

Figure 4 shows the motion velocity/acceleration

curved surface of expert and beginner in crawl

motion. They are formed by the joint trajectories of

right hand, right elbow, and right shoulder. Table 2

shows the maximum velocity, the minimum

velocity, the maximum acceleration, the minimum

acceleration, and timing of maximum velocity or

acceleration in crawl motion. From Figure 4, the

expert has the fastest in the downswing of right arm

(enter into a water), beginner has the fastest in

upswing of right arm (go out from a water). Expert’s

velocity is faster than the beginner’s velocity from

Table 2. Acceleration results are also. In the expert’s

opinion, he needs to push the water; he needs to

concentrate in enter to water.

3.3 Defend Motion in Karate

Figure 5 shows the motion velocity/acceleration

curved surface of expert and beginner in karate

defense motion. They are formed by the joint

trajectories of right hand, right elbow, and right

shoulder. Table 3 shows the maximum velocity, the

minimum velocity, the maximum acceleration, the

minimum acceleration, and timing of maximum

velocity or acceleration in defense motion. From

Figure 5, the expert has the fastest in the upswing of

both arms (defense start), beginner has the fastest in

the downswing of both arms (defense release).

Expert’s velocity is slower than the beginner’s

velocity from Table 3. In the expert’s opinion, the

arm trajectory is short, and he needs to concentrate

on stop arm (defend start); to accelerate the arm in

the upswing.

(a) Expert motion

(b) Beginner motion

(d) Velocity of beginner

(e) Acceleration of expert

(f) Acceleration of beginner

Figure 2: Motion velocity/acceleration curved surface in

darts throw.

Table 1: Maximum/minimum velocity/acceleration, timing

in darts throw.

Expert Beginner

max

3.30 2.43

min

0.00 0.00

max

27.98 29.60

min

-19.13 -9.84

Position Release Start

Suggestion of Motion Velocity/Acceleration Curved Surface

491

(a) Velocity of expert motion

(b) Acceleration of expert motion

(d) Acceleration of expert motion

Figure 3: Numerical analysis of velocity/acceleration.

4 CONCLUSION

We suggest the motion velocity curved surface and

the motion acceleration curved surface, and the

surfaces are investigated using 3 sports motion

samples for the validity. The motion velocity/

acceleration surface is expressed by time derivative,

(a) Expert motion

(b) Beginner motion

(d) Velocity of beginner

(e) Acceleration of expert

(f) Acceleration of beginner

Figure 4: Motion velocity/acceleration curved surface in

swimming crawl.

Table 2: Maximum/minimum velocity/acceleration, timing

in swimming crawl.

the express method is the RGB color gradation.

After that, the differences of expert and beginner

motion are visualized using velocity/ acceleration

from the surfaces. In this result, the difference is the

maximum velocity timing, and maximum

velocity/acceleration value. So, we can evaluate the

difference from velocity, acceleration, and timing. In

future work, we adopt the other motion samples

using the surface, and suggest the motion force

curved surface.

ACKNOWLEDGEMENTS

This work was supported by JSPS Grant-in-Aid for

Young Scientists (B) Grant Number JP16K16328.

Expert Beginner

max

6.02 5.73

min

0.00 0.00

max

64.96 47.85

min

-68.40 -85.52

Position Downswing Upswing

ICINCO 2017 - 14th International Conference on Informatics in Control, Automation and Robotics

492

(a) Expert motion

(b) Beginner motion

(d) Velocity of beginner

(e) Acceleration of expert

(f) Acceleration of beginner

Figure 5: Motion velocity/acceleration curved surface in

karate defence.

Table 3: Maximum/minimum velocity/acceleration, timing

in karate defence.

REFERENCES

Yoshitaka Fujino， Kousei Inoue， Masao Kikkawa，

Emi Nishina, and Tsuneo Ymamada. Sport

Onomatopoeia Data Base for Motor Learning, Japan

Society for Educational Technology, Vol.29, pp.5-8,

2005, (in Japanese)

Taki, T., Hasegawa, J. I., & Fukumura, T. (1996,

September). Development of motion analysis system

for quantitative evaluation of teamwork in soccer

games. In Image Processing, 1996. Proceedings,

International Conference on Vol. 3, pp. 815-818.

IEEE.

Kaoru Mitsuhashi, Hiroshi Hashimoto, and Yasuhiro

Ohyama, 2014. The Curved Surface Visualization of

the Expert Behavior for Skill Transfer Using

Microsoft Kinect, 11th International Conference on

Informatics in Control, Automation and Robotics

(ICINCO 2014), September1-3, Wien, Austria, 2014,

pp.550-555

Mitsuki Suneya, Masaki Kutsuna, Kaoru Mitsuhashi,

Yasuhiro Ohyama, and Hiroshi Hashimoto, 2014. The

Curved Surface Visualization of Behaviors for Skill

Inheriting using RGBD camera. SICE System

Integration 2014 (SI2014), December 14-17, Tokyo,

Japan. Pp.67-71, (in Japanese)

Kaoru Mitsuhashi, Hiroshi Hashimoto, and Yasuhiro

Ohyama, 2015. Motion Curved Surface Analysis and

Composite for Skill Succession using RGBD Camera,

12th International Conference on Informatics in

Control, Automation and Robotics (ICINCO 2015),

July21-23, France, 2015, pp.406-413

Kaoru MITSUHASHI, Hiroshi HASHIMOTO and

Yasuhiro OHYAMA, “Skill Level Evaluation of

Motion Curved Surface Character”, 13th International

Conference on Informatics in Control, Automation and

Robotics (ICINCO 2016), July 29-31, Lisbon, Portugal,

2016, pp.499-504

Expert Beginner

max

1.42 4.02

min

0.00 0.00

max

4.12 22.06

min

-3.30 -24.45

Position Upswing Downswing

Suggestion of Motion Velocity/Acceleration Curved Surface

493