Whole-Body Conditioned Egocentric Video Prediction

Jul 11, 2026

Berkeley AI Research has introduced PEVA, a model that predicts egocentric video frames based on human actions specified as 3D pose changes. The model can generate videos of atomic actions, simulate counterfactual scenarios, and support long video generation, addressing challenges in building world models for embodied agents with complex action spaces and egocentric perspectives.

Why it matters: This research advances world models for embodied AI by enabling video prediction conditioned on whole-body actions from an egocentric perspective.

Full story at: Berkeley AI Research ↗