Motion Laboratory

In football, there are multiple ways to kick the ball, and the ball's rotation axis and direction of rotation are different depending on the way of kicking. Therefore, in this study, I used motion data to compare the rotational motions of balls made by different kicking methods and to verify whether the balls rotate according to a theoretical principle. I measured the ball motion by four different kicking methods using an optical motion capture system. I calculated the rotation angle of each of frame from the initial frame based on each of three axes as the key feature of the ball. As a result, it was confirmed that in the case of the in-step kick method, the amplitude of the rotation angle with respect to the vertical axis was larger than with the other methods and the ball rotated longitudinally. However, in the case of the inside kick method with similar vertical rotation, the amplitude was small. Likewise, at the rotation angle with reference to the horizontal axis, it was confirmed that the amplitude of the outside kick method was larger than that of the in-step kick method and the ball was transversely rotated. However, in the case of the in-front kick method, the power of the kick was weak and thus the ball did not rotate so much.

On the side of shari caskets, even though masked dancers and musical performers are drawn, we can't quickly comprehend any detailed information. In this research, we developed interactive content using motion data and augmented reality (AR) to stimulate interest in shari caskets and promote their understanding. By recognizing the images of the dancers on the side of the replicas of a shari casket with a tablet's camera, 3DCG images of the corresponding dancers or shari caskets are overlaid and displayed. We added dance motion data by motion capture to the CG characters of the dancer, and we can see the same type of dance as reflected in the poses drawn on the casket. For interactive viewing, the CG character can also be operated with touch input or GUI buttons on the screen. To evaluate this content's utility, we conducted an evaluation experiment with ten subjects and confirmed its usefulness for stimulating interest and promoting the understanding of shari caskets.

In recent years, information technology has been adopted in museum exhibitions. However, it is not desirable to make visitors take a survey, so it is difficult to analyze how a given system is actually used. In this study, to clarify the requirements of museum exhibition systems and the types of knowledge used to appreciate an exhibition, I analyzed user behaviors from the log data of a system actually deployed at a museum. This system is used for an exhibit that offers experiences with sugoroku, a Japanese board game from the Edo period. Using the log data, I calculated the percentage of users who increased their scores. When users tried a dice-type device, they could intentionally manipulate their outcome in the game, so the percentage of people who could win was higher than that of those not using it. For a system that allows users to browse images of fountain pens using a head-mounted display (HMD), I again used the log data to calculate the number of browsing times per fountain pen. When the users wore the HMD, fountain pens in the center of the screen were often browsed regardless of their features. From these observations, we found that the results of a game could be changed through the interface, and when viewing a specific item among many exhibited items, the centrally positioned one is often the most browsed.

In such a group dance as Yosakoi, how to combine formations on stage is critically different from personal dance. In this research, to simulate the Yosakoi group dance, I developed a group simulation system using motion data. To represent the complicated formation movements commonly seen in Yosakoi, this system combines two formations or separates one formation from another formation. Since the leader of virtual dancers is often replaced in the transitions, the other virtual dancers are efficiently arranged on the right and left alternately based on the selected leader. Using the Yosakoi choreography that was acquired using motion capture, users can simulate the created group dance with 3DCG. To evaluate the utility of this system, I conducted an evaluation experiment with ten experienced Yosakoi dancers and confirmed that my system can be utilized as a creative tool for group dance.

This research aims to clarify the effects on the body of each pedaling form used in pedaling a road bike. I captured three pedaling forms by motion capture. The knee and ankle angles viewed from the side were calculated using markers attached to the human body. The foot angle and the rotation angle of the waist viewed from the top were calculated using markers attached to the road bike. Using the calculated feature values, I compared the pedaling forms among the normal sitting form, the standing form, and the sitting form while applying full power. As a result, in case of the sitting form with full power, the average of the maximum and minimum values of the right foot angle and the right knee angle, as well as the average amplitude of the right ankle angle viewed from the side, became larger than in the normal sitting form. In the full-power sitting form, I found that disturbance easily occurs in the movement of the entire lower body. In the case of the standing form, the average of the maximum and minimum values of both feet angles, as well as the average amplitude of the rotation angle of the waist, became lager than in the normal sitting form. The standing form was liable to disrupt the foot movement, but this should be expected because this form turned out to involve pedaling while twisting the waist greatly to the left and right.

Bouldering is a type of free climbing, and it's a very hard and dangerous sport. Only a limited number of facilities for playing this sport are available, so there are still few experienced participants. I developed a virtual reality simulator using a head-mounted display and its controllers to present a physical sense of bouldering and thus stimulate the interest of inexperienced people. This system allows users to control a virtual character's viewpoint and both of the character's hands with Oculus Rift and Oculus Touch. The viewpoint is changed by placing the hands on arranged protrusions called holds in the virtual world and moving them after pushing a button. This allows the wall to be climbed virtually using only upper-body motions. This system was implemented with such functions as reflecting the length of the user's arms in the virtual character, to increase the sense of immersion for users, and randomly arranging holds to allow users to climb the walls by various routes. From the questionnaire results of 31 users, about 80% responded positively to questions about operability, reproducibility of the arm movement, enjoyment, reproducibility of wall climbing, and increased interest in the sport. A suggestion for improvement was the need to bring this system closer to reality.

Since traffic accidents with road bicycles are increasing, we developed a road bike simulator with which bicyclists virtually experience traffic accidents. In this system, the user wears a HMD, a gesture controller and an acceleration sensor and rides on an actual bicycle within a VR space. The bicycle's braking action is detected by the myoelectric potential from a gesture controller called a MYO that is attached to the rider's forearm. By getting its orientation with a gyroscope, we can alter the bicycle's traveling direction. We changed the vehicle's moving speed in the VR space based on the number of pedal rotations per minute from the acceleration sensor attached to the leg. With our system, users experienced three types of serious accidents on road bikes: a left turn collision, being rear-ended, and sudden braking. We conducted an evaluation with a simulator that was used by five road bikers. Most subjects believed that their awareness of accidents was improved by experiencing accidents in VR.

As birthrates continue to decline and populations continue to age, muscular strength is declining in various ages due to fewer opportunities to exercise. In this research, I developed a system that enables easy enjoyment of exercise through games for continuous exercise indoors. In this system, the rehabilitation exercise that is performed by the user is acquired by a Kinect. This system estimates postures from the position of the acquired skeleton of the users and recognizes their motions. In a baby-chick breeding game, users can strengthen the muscles of their lower limbs by bending and stretching their knee, and flexing their shoulder joints by arm lifting. In a mini-obstacle game, users can keep walking by stamping and strengthening their entire leg muscles by jumping. In a sky-jumping mini game, users can strengthen their trunk and their lower limbs by one-leg standing. The baby-chick character and the background image are changed by the user motions so that users can exercise for a long time. This system allows them to decide their own exercise goals so that they achieve them. By recording the results, this system allows users to experience accomplishment and competitiveness. To evaluate the utility of this system, 16 subjects experienced it and answered questionnaires. 90% of my subjects effectively exercised and enjoyed the game.

In this basic research aimed at virtually drawing pictures on real objects, I developed a system that superimposes and displays hand-drawn information using smart glasses. In this system, the color of a pen's tip is detected and tracked, and lines are drawn on the traced trajectory. The drawing results can be viewed through smart glasses. In order to separate the drawn lines, the system switches between active and inactive locus drawing based on the distance from the camera at the pen tip. When the position of the pen is too close to the camera within a certain threshold, no lines are drawn. Even when the camera moves after drawing, the drawn information remains displayed and superimposed on the drawing surface. In order to confirm the operability of this system and the usefulness of the concept, I conducted an evaluation experiment using ten people. As a result, I confirmed the usefulness of my concept for virtually drawing pictures onto real objects in real space; however, there still remains the problem of the system's operability.