Learning Signed Distance Field for Multi-view Surface Reconstruction
This work addresses the problem of reconstructing complex scene topologies in computer vision, offering improved robustness for applications like 3D modeling, but it is incremental as it builds on existing implicit neural representation frameworks.
The paper tackles multi-view surface reconstruction for complex and concave objects without requiring clean object masks, achieving better mesh reconstruction in wide open scenes compared to previous state-of-the-art methods on datasets like DTU, EPFL, and Tanks and Temples.
Recent works on implicit neural representations have shown promising results for multi-view surface reconstruction. However, most approaches are limited to relatively simple geometries and usually require clean object masks for reconstructing complex and concave objects. In this work, we introduce a novel neural surface reconstruction framework that leverages the knowledge of stereo matching and feature consistency to optimize the implicit surface representation. More specifically, we apply a signed distance field (SDF) and a surface light field to represent the scene geometry and appearance respectively. The SDF is directly supervised by geometry from stereo matching, and is refined by optimizing the multi-view feature consistency and the fidelity of rendered images. Our method is able to improve the robustness of geometry estimation and support reconstruction of complex scene topologies. Extensive experiments have been conducted on DTU, EPFL and Tanks and Temples datasets. Compared to previous state-of-the-art methods, our method achieves better mesh reconstruction in wide open scenes without masks as input.