Seeing the Wind from a Falling Leaf
This work addresses the challenge of modeling physical interactions from videos for applications in computer vision and physics-based video generation, representing a novel but domain-specific advancement.
The paper tackles the problem of recovering invisible physical forces, such as wind fields, from visual observations of object motions in videos, using an end-to-end differentiable inverse graphics framework, and demonstrates its ability to infer plausible force fields on synthetic and real-world scenarios.
A longstanding goal in computer vision is to model motions from videos, while the representations behind motions, i.e. the invisible physical interactions that cause objects to deform and move, remain largely unexplored. In this paper, we study how to recover the invisible forces from visual observations, e.g., estimating the wind field by observing a leaf falling to the ground. Our key innovation is an end-to-end differentiable inverse graphics framework, which jointly models object geometry, physical properties, and interactions directly from videos. Through backpropagation, our approach enables the recovery of force representations from object motions. We validate our method on both synthetic and real-world scenarios, and the results demonstrate its ability to infer plausible force fields from videos. Furthermore, we show the potential applications of our approach, including physics-based video generation and editing. We hope our approach sheds light on understanding and modeling the physical process behind pixels, bridging the gap between vision and physics. Please check more video results in our \href{https://chaoren2357.github.io/seeingthewind/}{project page}.