Scene Flow
Scene flow estimation aims to recover per-point motion from two adjacent LiDAR scans. However, in real-world applications such as autonomous driving, points rarely move independently of others, especially for nearby points belonging to the same object, which often share the same motion. Incorporating this locally rigid motion constraint has been a key challenge. We propose data and compute-efficient scene flow models that are free from human annotation during training and capable of real-time inference at test time.
CVL people: Yancong Lin
