PV-RAFT: Point-Voxel Correlation Fields for Scene Flow Estimation of Point Clouds
This work provides a significant improvement in scene flow estimation for autonomous driving and robotics applications, where accurate motion understanding of 3D environments is crucial.
This paper addresses the challenge of scene flow estimation from irregular and unordered point clouds by proposing PV-RAFT, a method that captures both local and long-range dependencies of point pairs. It achieves state-of-the-art performance on the FlyingThings3D and KITTI Scene Flow 2015 datasets, outperforming existing methods by remarkable margins.
In this paper, we propose a Point-Voxel Recurrent All-Pairs Field Transforms (PV-RAFT) method to estimate scene flow from point clouds. Since point clouds are irregular and unordered, it is challenging to efficiently extract features from all-pairs fields in the 3D space, where all-pairs correlations play important roles in scene flow estimation. To tackle this problem, we present point-voxel correlation fields, which capture both local and long-range dependencies of point pairs. To capture point-based correlations, we adopt the K-Nearest Neighbors search that preserves fine-grained information in the local region. By voxelizing point clouds in a multi-scale manner, we construct pyramid correlation voxels to model long-range correspondences. Integrating these two types of correlations, our PV-RAFT makes use of all-pairs relations to handle both small and large displacements. We evaluate the proposed method on the FlyingThings3D and KITTI Scene Flow 2015 datasets. Experimental results show that PV-RAFT outperforms state-of-the-art methods by remarkable margins.