RMS-FlowNet: Efficient and Robust Multi-Scale Scene Flow Estimation for Large-Scale Point Clouds
This addresses the challenge of processing dense point clouds in applications like autonomous driving, though it is incremental as it builds on existing learning-based methods with specific improvements.
The paper tackles the problem of efficient and robust scene flow estimation for large-scale point clouds by proposing RMS-FlowNet, which uses random sampling and a novel flow embedding to achieve faster prediction than state-of-the-art methods while handling over 250K points at once.
The proposed RMS-FlowNet is a novel end-to-end learning-based architecture for accurate and efficient scene flow estimation which can operate on point clouds of high density. For hierarchical scene flow estimation, the existing methods depend on either expensive Farthest-Point-Sampling (FPS) or structure-based scaling which decrease their ability to handle a large number of points. Unlike these methods, we base our fully supervised architecture on Random-Sampling (RS) for multiscale scene flow prediction. To this end, we propose a novel flow embedding design which can predict more robust scene flow in conjunction with RS. Exhibiting high accuracy, our RMS-FlowNet provides a faster prediction than state-of-the-art methods and works efficiently on consecutive dense point clouds of more than 250K points at once. Our comprehensive experiments verify the accuracy of RMS-FlowNet on the established FlyingThings3D data set with different point cloud densities and validate our design choices. Additionally, we show that our model presents a competitive ability to generalize towards the real-world scenes of KITTI data set without fine-tuning.