AI RESEARCH
Motion Dreamer: Boundary Conditional Motion Reasoning for Physically Coherent Video Generation
arXiv CS.AI
•
ArXi:2412.00547v4 Announce Type: replace-cross Recent advances in video generation have shown promise for generating future scenarios, critical for planning and control in autonomous driving and embodied intelligence. However, real-world applications demand than visually plausible predictions; they require reasoning about object motions based on explicitly defined boundary conditions, such as initial scene image and partial object motion. We term this capability Boundary Conditional Motion Reasoning.