Development of Markerless Motion Capture Method adaptive to the environment with obstacles

2017 Fiscal Year Final Research Report

• Project/Area Number: 15H03076
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Sports science
• Research Institution: Kogakuin University
• Principal Investigator: Kiriyama Yoshimori 工学院大学, 工学部, 准教授 (30383722)
• Research Collaborator: SEZAKO Kanta
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: マーカレスモーションキャプチャ / 動作計測 / 遮蔽物存在環境下

Outline of Final Research Achievements

Markerless Motion capture Method (MMC) is thought to be one of the motion capture method in the next generation. In this study, we developed the method to use MMC under the environment where there are some obstacles. Stereo vision was used to make a higher precision visual hull. Also, Joint surface model was developed to estimate the distal segments based on the deformation of the joint surface. We performed gait analysis under the condition to hide the body segments. Our method was restricted because the camera resolution was too low to use the proposed method, but we used a novel evaluation to match the visual hull and a subject model. As a result, we were able to use MMC to analyze human gait even when 60% of a body segment was hidden.

Free Research Field

生体力学，医用工学，スポーツ工学