Self-Supervised Learning of Non-Rigid Residual Flow and Ego-Motion

• Published in: 2020 International Conference on 3D Vision (3DV) pp. 150 - 159
• Main Authors: Tishchenko, Ivan, Lombardi, Sandro, Oswald, Martin R., Pollefeys, Marc
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2020
• Subjects: Cameras; Deep learning; Estimation; Lattices; Scene Flow; Self Supervised Learning; Three-dimensional displays; Training; Transforms
• ISSN: 2475-7888
• DOI: 10.1109/3DV50981.2020.00025

Most of the current scene flow methods choose to model scene flow as a per point translation vector without differentiating between static and dynamic components of 3D motion. In this work we present an alternative method for end-to-end scene flow learning by joint estimation of non-rigid residual flow and ego-motion flow for dynamic 3D scenes. We propose to learn the relative rigid transformation from a pair of point clouds followed by an iterative refinement. We then learn the non-rigid flow from transformed inputs with the deducted rigid part of the flow. Furthermore, we extend the supervised framework with self-supervisory signals based on the temporal consistency property of a point cloud sequence. Our solution allows both training in a supervised mode complemented by self-supervisory loss terms as well as training in a fully self-supervised mode. We demonstrate that decomposition of scene flow into non-rigid flow and ego-motion flow along with an introduction of the self-supervisory signals allowed us to outperform the current state-of-the-art supervised methods.