Smol-GS: Compact Representations for Abstract 3D Gaussian Splatting
This work addresses the need for efficient 3D scene representations for applications like navigation and planning, though it appears incremental as it builds on existing 3D Gaussian Splatting methods.
The paper tackled the problem of compressing 3D Gaussian Splatting representations by introducing Smol-GS, a method that learns compact encodings integrating spatial and semantic information, achieving state-of-the-art compression on standard benchmarks while maintaining high rendering quality.
We present Smol-GS, a novel method for learning compact representations for 3D Gaussian Splatting (3DGS). Our approach learns highly efficient encodings in 3D space that integrate both spatial and semantic information. The model captures the coordinates of the splats through a recursive voxel hierarchy, while splat-wise features store abstracted cues, including color, opacity, transformation, and material properties. This design allows the model to compress 3D scenes by orders of magnitude without loss of flexibility. Smol-GS achieves state-of-the-art compression on standard benchmarks while maintaining high rendering quality. Beyond visual fidelity, the discrete representations could potentially serve as a foundation for downstream tasks such as navigation, planning, and broader 3D scene understanding.