Tag
This paper proposes COVER, a training-free method for converting 3D assets into sparse panoramic RGB-D-pose data with complete scene coverage and low redundancy, and introduces the CM-EVS dataset containing 36,373 curated frames from indoor and outdoor scenes.
This paper introduces HERMES++, a unified driving world model that integrates 3D scene understanding and future geometry prediction using BEV representation, LLM-enhanced queries, and joint geometric optimization.