DGNS: Deformable Gaussian Splatting and Dynamic Neural Surface for Monocular Dynamic 3D Reconstruction
This addresses the problem of reconstructing dynamic scenes from single-camera videos for real-world applications, representing an incremental improvement over existing methods.
The paper tackles dynamic 3D reconstruction from monocular video by introducing DGNS, a hybrid framework that integrates deformable Gaussian splatting and dynamic neural surfaces, achieving state-of-the-art performance in 3D reconstruction and competitive results in novel-view synthesis on public datasets.
Dynamic scene reconstruction from monocular video is essential for real-world applications. We introduce DGNS, a hybrid framework integrating \underline{D}eformable \underline{G}aussian Splatting and Dynamic \underline{N}eural \underline{S}urfaces, effectively addressing dynamic novel-view synthesis and 3D geometry reconstruction simultaneously. During training, depth maps generated by the deformable Gaussian splatting module guide the ray sampling for faster processing and provide depth supervision within the dynamic neural surface module to improve geometry reconstruction. Conversely, the dynamic neural surface directs the distribution of Gaussian primitives around the surface, enhancing rendering quality. In addition, we propose a depth-filtering approach to further refine depth supervision. Extensive experiments conducted on public datasets demonstrate that DGNS achieves state-of-the-art performance in 3D reconstruction, along with competitive results in novel-view synthesis.