Active observation and behavior recognition for the assistance of handicapped people

2016 Fiscal Year Final Research Report

• Project/Area Number: 26330195
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Perceptual information processing
• Research Institution: Wakayama University
• Principal Investigator: Chen Qian 和歌山大学, システム工学部, 准教授 (70263233)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 円パターン / 能動的カメラ / カメラ制御 / カメラ較正 / 追跡 / 環境認識 / 人物の検出

Outline of Final Research Achievements

In this research, we used circle pattern and iterative voting to detect floor plane with stairs and to estimate the pose and the position of the camera. We used two active camera to observe a person cooperatively to enable the observation of a person contentiously when he walks around among multiple rooms. We also used inferred camera to detect faces in night scenes. In order to obtain clear image of people, we developed cepstrum-based algorithms to detect blurring information and a new de-blur algorithm based on it.

Free Research Field

コンピュータビジョン