Assistant Professor, Brown University
2 papers at NeurIPS 2025
We propose a unified and generalizable framework for object state recognition in video data.
Video generalization models can learn physics-based controls from synthetic data and generalize this control across diverse geometries, settings, and materials.
