Locality-aware Gaussian Compression for Fast and High-quality Rendering
This work addresses storage and speed bottlenecks for real-time, high-quality 3D rendering in applications like VR and gaming, representing an incremental improvement over prior compression techniques.
The paper tackles the problem of high storage and slow rendering in 3D Gaussian Splatting by introducing LocoGS, a locality-aware framework that compresses 3D Gaussian representations, achieving up to 96.6× storage reduction and 2.4× faster rendering while improving quality over existing methods.
We present LocoGS, a locality-aware 3D Gaussian Splatting (3DGS) framework that exploits the spatial coherence of 3D Gaussians for compact modeling of volumetric scenes. To this end, we first analyze the local coherence of 3D Gaussian attributes, and propose a novel locality-aware 3D Gaussian representation that effectively encodes locally-coherent Gaussian attributes using a neural field representation with a minimal storage requirement. On top of the novel representation, LocoGS is carefully designed with additional components such as dense initialization, an adaptive spherical harmonics bandwidth scheme and different encoding schemes for different Gaussian attributes to maximize compression performance. Experimental results demonstrate that our approach outperforms the rendering quality of existing compact Gaussian representations for representative real-world 3D datasets while achieving from 54.6$\times$ to 96.6$\times$ compressed storage size and from 2.1$\times$ to 2.4$\times$ rendering speed than 3DGS. Even our approach also demonstrates an averaged 2.4$\times$ higher rendering speed than the state-of-the-art compression method with comparable compression performance.