DELFlow: Dense Efficient Learning of Scene Flow for Large-Scale Point Clouds

TL;DR

DELFlow introduces a dense 2D grid representation for point clouds, enabling efficient and effective scene flow estimation by preserving points and improving feature fusion, outperforming previous methods on standard datasets.

Contribution

The paper proposes a novel dense 2D point cloud representation and a warping projection technique to enhance efficiency and accuracy in scene flow estimation.

Findings

Outperforms prior methods on FlyingThings3D and KITTI datasets.

Significantly improves efficiency by avoiding costly sampling operations.

Effectively preserves scene information with dense point representation.

Abstract

Point clouds are naturally sparse, while image pixels are dense. The inconsistency limits feature fusion from both modalities for point-wise scene flow estimation. Previous methods rarely predict scene flow from the entire point clouds of the scene with one-time inference due to the memory inefficiency and heavy overhead from distance calculation and sorting involved in commonly used farthest point sampling, KNN, and ball query algorithms for local feature aggregation. To mitigate these issues in scene flow learning, we regularize raw points to a dense format by storing 3D coordinates in 2D grids. Unlike the sampling operation commonly used in existing works, the dense 2D representation 1) preserves most points in the given scene, 2) brings in a significant boost of efficiency, and 3) eliminates the density gap between points and pixels, allowing us to perform effective feature fusion. We also present a novel warping projection technique to alleviate the information loss problem resulting from the fact that multiple points could be mapped into one grid during projection when computing cost volume. Sufficient experiments demonstrate the efficiency and effectiveness of our method, outperforming the prior-arts on the FlyingThings3D and KITTI dataset.