PhD Thesis

Learning representations for human motion understanding

Work default illustration

Information

  • Started: 12/12/2019

Description

Vision-based analysis of humans and their actions is an active research area, with potential applications in robotics and VR/AR. This thesis will investigate new approaches for prediction and understanding of 3D human pose, shape, and motion, by leveraging novel geometric cues.

The work is under the scope of the following projects:

  • HuMoUR: Markerless 3D human motion understanding for adaptive robot behavior (web)