Improved 3D scene reconstruction using Gaussian spacing normalization
SplatAnything: Gaussian Spacing Normalization for Feed-forward 3DGS across Arbitrary Scenes
Computer Vision and Pattern Recognition
Summary
Reconstructing 3D scenes using existing methods can be tricky because the distances between key points called Gaussians change a lot depending on the scene size and viewpoint. This makes it hard for the system to learn the right shapes consistently. The authors found a way to adjust these key points by normalizing how spaced out they are across the whole scene, making the system's predictions more stable. They tested their method on many different scenes and camera views and showed it works better than previous approaches, especially when the scene size or viewpoint changes a lot.
What this means in practice
- •For 3d graphics developers: Create more accurate and stable 3D scene reconstructions when viewpoints or scene scales vary significantly.
- •For robot navigation teams: Improve environment mapping consistency for robots operating across different scales and viewing conditions.
Authors
Hao Yu, Zhenwei Wang, Yongzhen Hu, Yuning Gong, Junta Wu, Xiaowei Zhou, Hujun Bao, Tengfei Wang, Chunchao Guo, Sida Peng
Abstract
We identify a key limitation of existing feed-forward 3DGS methods: Gaussian attributes are typically predicted in a radial-distance-normalized space that canonicalizes a global radial statistic of the reconstructed geometry, while the spacing between neighboring Gaussian centers can still vary substantially across scene scales and camera configurations. This mismatch induces systematic shifts in Gaussian-scale targets, leading to ambiguous supervision. To address this, we propose SplatAnything with Gaussian Spacing Normalization (GSNorm), which canonicalizes a robust scene-level statistic of Gaussian spacing to provide a more consistent spatial reference for Gaussian prediction. To rigorously assess our method, we further establish a systematic benchmark spanning diverse datasets, input sparsity levels, and challenging extrapolated viewpoints. Extensive experiments demonstrate that SplatAnything achieves state-of-the-art performance, with particularly strong gains under substantial variations in scene scale and viewpoint.