project

Articulated human body tracking through clutter


Provided by   Curtin University

Research Project

Researchers: Patrick Peursum (Managed by)

Brief description This project addresses the problem of markerless tracking of a human in full 3D with a high-dimensional (29D) body model. Most work in this area has been focused on achieving accurate tracking in order to replace marker-based motion capture, but do so at the cost of relying on relatively clean observing conditions. This project takes a different perspective, proposing a body-tracking model that is explicitly designed to handle real-world conditions such as occlusions by scene objects, failure recovery, long-term tracking, auto-initialisation, generalisation to different people and integration with action recognition. As part of this, ground-truth information of the person’s posture is required to evaluate tracking accuracy.

Click to explore relationships graph
Viewed: [[ro.stat.viewed]]