Pattern recognition using graph signal processing for large-scale time-sequence data

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K12061
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Perceptual information processing
• Research Institution: Tokyo Institute of Technology
• Principal Investigator: Koichi Shinoda 東京工業大学, 情報理工学院, 教授 (10343097)
• Co-Investigator(Kenkyū-buntansha): 井上 中順 東京工業大学, 情報理工学院, 助教 (10733397)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 動作認識 / グラフ信号処理 / 深度カメラ

Outline of Final Research Achievements

We have developed an action recognition method from RGB-D camera inputs. This method uses a time sequence of human skeleton as an input. Every frame it extracts features by using spectral graph wavelet transform. Then the features are pooled in a hierarchical way in the time axis. This method achieved the state-of-the-art in multi-view action recognition.

Free Research Field

統計的パターン認識