Human Body, Hands, and Activities from Egocentric and Multi-view Cameras

Welcome!

Egocentric perception of humans is a key feature for rapidly developing AR/VR glasses, human-computer interactions, and other assistive robotics. Deploying these applications requires considering the human in a 3D world. Indeed, perceiving and understanding humans from a third-person view is a long standing research topic. Human understanding from an egocentric (or first-person) perspective, however, is significantly less explored and encompasses its own unique challenges, such as analysing and predicting the motion, body/hand reconstruction and action categories with other humans and the environment, all from strongly truncated and/or motion-blurred first-person views. The first edition of this workshop aims to gather researchers working on egocentric body pose, hand pose and 3D activity recognition and associated applications at ECCV 2022.

The workshop includes four challenging benchmarks, and we accept extended abstracts (2-4 pages) about:

• challenge participations

• related topics on egocentric body pose, hand pose and 3D activity recognition