Motion History Volumes for Free Viewpoint Action Recognition
Year: 2005
Type: conference-paper
Abstract: Action recognition is an important and challenging topic in computer vision, with many important applications including video surveillance, automated cinematography and understanding of social interaction. Yet, most current work in gesture or action interpretation remains rooted in viewdependent representations. This paper introduces Motion History Volumes (MHV) as a free-viewpoint representation for human actions in the case of multiple calibrated, and background-subtracted, video cameras. We present algorithms for computing, aligning and comparing MHVs of different actions performed by different people in a variety of viewpoints. Alignment and comparisons are performed efficiently using Fourier transforms in cylindrical coordinates around the vertical axis. Preliminary results indicate that this representation can be used to learn and recognize basic human action classes, independently of gender, body size and viewpoint.
Source: HAL
Authors: Daniel Weinland, Rémi Ronfard, Edmond Boyer
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W66091190