3R-GS: Best Practice in Optimizing Camera Poses Along with 3DGS
This addresses a bottleneck in neural rendering for 3D reconstruction, offering incremental improvements in robustness and precision for challenging scenes.
The paper tackles the problem of 3D Gaussian Splatting's dependence on accurate camera poses by proposing 3R-GS, a framework that jointly optimizes 3D Gaussians and camera parameters, resulting in high-quality novel view synthesis and precise camera pose estimation with computational efficiency.
3D Gaussian Splatting (3DGS) has revolutionized neural rendering with its efficiency and quality, but like many novel view synthesis methods, it heavily depends on accurate camera poses from Structure-from-Motion (SfM) systems. Although recent SfM pipelines have made impressive progress, questions remain about how to further improve both their robust performance in challenging conditions (e.g., textureless scenes) and the precision of camera parameter estimation simultaneously. We present 3R-GS, a 3D Gaussian Splatting framework that bridges this gap by jointly optimizing 3D Gaussians and camera parameters from large reconstruction priors MASt3R-SfM. We note that naively performing joint 3D Gaussian and camera optimization faces two challenges: the sensitivity to the quality of SfM initialization, and its limited capacity for global optimization, leading to suboptimal reconstruction results. Our 3R-GS, overcomes these issues by incorporating optimized practices, enabling robust scene reconstruction even with imperfect camera registration. Extensive experiments demonstrate that 3R-GS delivers high-quality novel view synthesis and precise camera pose estimation while remaining computationally efficient. Project page: https://zsh523.github.io/3R-GS/