CaT-GS: Efficient 3DGS Rendering for Large Scale Scenes via Inter-frame Caching and Tile Scheduling

2026-07-20

AI Summary

This work addresses the rendering efficiency bottleneck of 3D Gaussian Splatting (3DGS) in large-scale scenes, where tile-based rasterization incurs substantial computational overhead. Existing approaches either require retraining or optimize along a single dimension. To overcome these limitations, we propose CaT-GS, an efficient rendering pipeline that introduces, for the first time, cross-frame speculative preprocessing and caching to eliminate inter-frame redundancy. Additionally, we design a dedicated rasterization kernel with tile-level load balancing to mitigate view-dependent occlusion and uneven workload distribution. Our method requires no retraining, significantly improves GPU utilization, and achieves up to a 10× speedup over the original 3DGS while preserving high visual fidelity—surpassing the current state-of-the-art by 70% and establishing a new benchmark for real-time rendering of large-scale scenes.

Abstract

Recent breakthroughs in 3D Gaussian Splatting (3DGS) have advanced neural rendering with high fidelity and speed. However, its performance degrades significantly in large-scale scenes due to the computational burden of tile-based rasterization. Existing optimization efforts either require costly scene re-training or focus on narrow aspects of the pipeline, overlooking critical inefficiencies in real-world deployments. Through a comprehensive analysis, we identify three primary sources of redundancy and low GPU utilization: redundant inter-frame pre-processing, viewpoint-based occlusion redundancy, and severe tile-level load imbalance. To address these issues, we propose CaT-GS, a novel and efficient 3DGS rendering pipeline. CaT-GS introduces a speculative multi-frame preprocessing method to eliminate redundant computations across consecutive frames, and an inter-frame caching mechanism to eliminate viewpoint redundant rendering stages. Furthermore, it refactors rasterization tasks with a dedicated kernel to mitigate tile load imbalance, significantly boosting GPU utilization. Extensive experiments demonstrate that CaT-GS achieves a speedup of up to 10 times over the original 3DGS and up to 70% over previous state-of-the-art methods, establishing a new benchmark for high-fidelity, real-time rendering of large-scale scenes.

Problem

• 3D Gaussian Splatting
• large-scale scenes
• rendering efficiency
• tile-level load imbalance
• inter-frame redundancy

Innovation

• 3D Gaussian Splatting
• inter-frame caching
• tile scheduling
• real-time rendering
• GPU utilization