InHabit: Leveraging Image Foundation Models for Scalable 3D Human Placement 事件

InHabit: Leveraging Image Foundation Models for Scalable 3D Human Placement arXiv:2604.19673v2 Announce Type: replace Abstract: Training embodied agents to understand 3D scenes as humans do requires large-scale data of people meaningfully interacting with diverse environments, yet such data is scarce. Real-world capture is costly and limited to controlled settings, while existing synthetic datasets rely on simple geometric heuristics, ignoring rich scene context. In contrast, 2D foundation mode