Kinect image processing by CNN algorithm for gait recognition
Conference: CNNA 2016 - 15th International Workshop on Cellular Nanoscale Networks and their Applications
08/23/2016 - 08/25/2016 at Dresden, Deutschland
Proceedings: CNNA 2016
Pages: 2Language: englishTyp: PDF
Personal VDE Members are entitled to a 10% discount on this title
Authors:
Leyden, Kevin; Schmiedeler, James (Department of Aerospace and Mechanical Engineering, University of Notre Dame, Notre Dame, IN 46556, USA)
Koller, Miklos; Cserey, Gyoergy (Faculty of Information Technology and Bionics, Pazmany Peter Catholic University, 1083 Budapest, Hungary)
Niemier, Michael (Department of Computer Science and Engineering, University of Notre Dame, Notre Dame, IN 46556, USA)
Abstract:
The CNN algorithm in this paper is a link between data collection with the Microsoft Kinect and gait recognition by machine learning. It extracts body points and angles from grayscale depth images with anthropometric information, bypassing the need for the Kinect’s built-in mock skeleton. The trajectories of these points and angles are suitable inputs for identifying people according to their gaits.