Scale-GS: Efficient Scalable Gaussian Splatting via Redundancy-filtering Training on Streaming Content

2025-08-29

AI Summary

To address the high computational cost and slow per-frame training in dynamic scenes caused by dense Gaussian primitives in 3D Gaussian Splatting (3DGS), this paper proposes a scalable hierarchical Gaussian lattice framework. Our method introduces: (1) an anchor-based multi-scale Gaussian hierarchy enabling coarse-to-fine adaptive activation; (2) a synergistic optimization mechanism integrating deformation modeling with generative updates to enhance large-motion representation; and (3) bidirectional adaptive masking, redundant region filtering, and streaming data processing to substantially reduce memory footprint and training overhead. Experiments demonstrate that our approach maintains high-fidelity rendering quality while accelerating training by 2.1–3.8× over state-of-the-art methods. To our knowledge, this is the first work achieving real-time, efficient, and scalable Gaussian rendering for dynamic scenes.

Technology Category

• Computer Vision: 3D Computer Vision
• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Learning with Manifolds

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

3D Gaussian Splatting (3DGS) enables high-fidelity real-time rendering, a key requirement for immersive applications. However, the extension of 3DGS to dynamic scenes remains limitations on the substantial data volume of dense Gaussians and the prolonged training time required for each frame. This paper presents M, a scalable Gaussian Splatting framework designed for efficient training in streaming tasks. Specifically, Gaussian spheres are hierarchically organized by scale within an anchor-based structure. Coarser-level Gaussians represent the low-resolution structure of the scene, while finer-level Gaussians, responsible for detailed high-fidelity rendering, are selectively activated by the coarser-level Gaussians. To further reduce computational overhead, we introduce a hybrid deformation and spawning strategy that models motion of inter-frame through Gaussian deformation and triggers Gaussian spawning to characterize wide-range motion. Additionally, a bidirectional adaptive masking mechanism enhances training efficiency by removing static regions and prioritizing informative viewpoints. Extensive experiments demonstrate that M~ achieves superior visual quality while significantly reducing training time compared to state-of-the-art methods.

Problem

• Efficient scalable Gaussian Splatting for streaming content
• Reducing computational overhead in dynamic scene rendering
• Minimizing training time while maintaining visual quality

Innovation

• Hierarchical anchor-based Gaussian organization
• Hybrid deformation and spawning strategy
• Bidirectional adaptive masking mechanism