EvoGS: Constructing Continuous-Layered Gaussian Splatting with Evolution Tree for Scalable 3D Streaming
This work addresses the scalability and quality transition problems in progressive 3D streaming for real-time adaptive streaming applications.
EvoGS introduces a continuous-layering representation for 3D Gaussian Splatting streaming, organized as an Evolution Tree with wavelet-inspired parent-child refinement, reducing splat redundancy from over 65% to under 25% and achieving up to 2.4× lower transmission payload and 5.5× lower GPU VRAM footprint compared to state-of-the-art baselines.
Streaming 3D Gaussian Splatting requires highly scalable, progressive representations. Existing progressive methods rely on \textit{discrete layering}, accumulating separate splat sets for each level of detail. This structural independence between layers inherently leads to error accumulation, severe splat redundancy, and uncontrolled quality transitions. We propose EvoGS, the first \textit{continuous-layering} representation. Organized as an Evolution Tree, EvoGS generates finer details via an explicit, wavelet-inspired parent-child refinement. This empowers child nodes to structurally correct ancestral errors, yield inherently sparse and highly compressible inter-layer signals. Extensive experiments show EvoGS eliminates splat redundancy from over 65\% to under 25\%. Compared to state-of-the-art baselines, it reduces transmission payload and GPU VRAM footprint by up to 2.4$\times$ and 5.5$\times$, respectively, and achieves smooth quality transitions optimal for real-time adaptive streaming. Project page: https://yuang-ian.github.io/evogs/