DEFLOW: Self-supervised 3D Motion Estimation of Debris Flow
This addresses the need for automated motion analysis in natural phenomena like debris flows, which is important for environmental monitoring and hazard assessment.
The paper tackles the problem of automated 3D motion estimation for debris flows, which lacks existing solutions, by proposing DEFLOW with a new dataset and achieving state-of-the-art optical flow and depth estimation on this dataset.
Existing work on scene flow estimation focuses on autonomous driving and mobile robotics, while automated solutions are lacking for motion in nature, such as that exhibited by debris flows. We propose DEFLOW, a model for 3D motion estimation of debris flows, together with a newly captured dataset. We adopt a novel multi-level sensor fusion architecture and self-supervision to incorporate the inductive biases of the scene. We further adopt a multi-frame temporal processing module to enable flow speed estimation over time. Our model achieves state-of-the-art optical flow and depth estimation on our dataset, and fully automates the motion estimation for debris flows. The source code and dataset are available at project page.