Bi-PointFlowNet: Bidirectional Learning for Point Cloud Based Scene Flow Estimation
This addresses the problem of limited accuracy and generality in scene flow estimation for computer vision applications, representing a novel method rather than an incremental improvement.
The paper tackled scene flow estimation by introducing a bidirectional flow embedding layer to learn features in both forward and backward directions, achieving a new state-of-the-art record with large margins on FlyingThings3D and KITTI benchmarks.
Scene flow estimation, which extracts point-wise motion between scenes, is becoming a crucial task in many computer vision tasks. However, all of the existing estimation methods utilize only the unidirectional features, restricting the accuracy and generality. This paper presents a novel scene flow estimation architecture using bidirectional flow embedding layers. The proposed bidirectional layer learns features along both forward and backward directions, enhancing the estimation performance. In addition, hierarchical feature extraction and warping improve the performance and reduce computational overhead. Experimental results show that the proposed architecture achieved a new state-of-the-art record by outperforming other approaches with large margin in both FlyingThings3D and KITTI benchmarks. Codes are available at https://github.com/cwc1260/BiFlow.