ICLR2025
Neural Eulerian Scene Flow Fields
Kyle Vedder, Neehar Peri, Ishan Khatri, Siyi Li, Eric Eaton, Mehmet Kemal Kocamaz, Yue Wang, Zhiding Yu, Deva Ramanan, Joachim Pehserl
Abstract
2 NVIDIA 3 Carnegie Mellon University (a) Small object motion extraction... (b) ...in diverse, dynamic scenes... (c) ...with emergent 3D point tracking behavior! Figure 1 : EulerFlow is able to capture the motion of small, fast moving objects with few lidar points, such a bird flying in front of an autonomous vehicle (Figure 1a ). EulerFlow's flexibility allows it to estimate scene flow for fast-moving table top objects without additional hyperparameter tuning (Figure 1b ). EulerFlow's ODE estimate exhibits emergent 3D point tracking behavior without explicit long-horizon supervision (Figure 1c ). Note that point clouds are shown in color for visualization purposes only; RGB is not used during optimization.