Multi-modal gait recognition in the wild and its application to criminal investigation

2017 Fiscal Year Final Research Report

• Project/Area Number: 15H01693
• Research Category: Grant-in-Aid for Scientific Research (A)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Perceptual information processing
• Research Institution: Osaka University
• Principal Investigator: YAGI Yasushi 大阪大学, 産業科学研究所, 教授 (60231643)
• Co-Investigator(Kenkyū-buntansha): 黒沢 健至 科学警察研究所, 法科学第二部, 室長 (80356170)
• Co-Investigator(Renkei-kenkyūsha): MAKIHARA Yasushi 大阪大学, 産業科学研究所, 准教授 (90403005) ; MURAMATSU Daigo 大阪大学, 産業科学研究所, 准教授 (00386624) ; KUROKI Kenro 科学警察研究所, 法科第二部, 研究員 (60392271) ; TSUCHIYA Ken'ichi 科学警察研究所, 法科第二部, 主任研究官 (90447920)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 歩容認証 / ながら歩行 / 計量学習 / データベース / アクセスコントロール

Outline of Final Research Achievements

We proposed a method of gait recognition in the wild (GRW) to effectively handle gait with some accompanied actions such as operating a smartphone. For this purpose, we constructed the world-largest gait database including approx. 60,000 subjects with carried objects, and then proposed a method of gait recognition robust against carrying status variations using joint intensity and spatial metric learning. Moreover, we developed an online walk-through multi-modal access control system, named, Gait Gate.

Free Research Field

視覚情報処理