Neural SDF Flow for 3D Reconstruction of Dynamic Scenes

In this paper, we tackle the problem of 3D reconstruction of dynamic scenes from multi-view videos. Previous dynamic scene reconstruction works either attempt to model the motion of 3D points in space, which constrains them to handle a single articulated object or require depth maps as input. By contrast, we propose to directly estimate the change of Signed Distance Function (SDF), namely SDF flow, of the dynamic scene. We show that the SDF flow captures the evolution of the scene surface. We further derive the mathematical relation between the SDF flow and the scene flow, which allows us to calculate the scene flow from the SDF flow analytically by solving linear equations. Our experiments on real-world multi-view video datasets show that our reconstructions are better than those of the state-of-the-art methods. Our code is available at https://github.com/ wei-mao-2019/SDFFlow.git . In this section, we first briefly introduce the neural radiance field and the SDF-based parameterization of the density (Section 3.1). We then describe our SDF flow to capture the dynamic scenes (Section 3.2). Lastly, we derive the mathematical relationship between the SDF flow and the scene flow (Section 3.3).