SeeLe: A Unified Acceleration Framework for Real-Time Gaussian Splatting

2025-03-07

AI Summary

To address the challenge of real-time 3D Gaussian Splatting (3DGS) rendering on mobile devices—constrained by limited GPU compute and memory capacity—this paper proposes an efficient mobile-oriented rendering framework. Our method introduces three key innovations: (1) a GPU-aware hybrid preprocessing pipeline that jointly optimizes view-dependent scene representations; (2) contribution-aware rasterization coupled with lightweight online Gaussian culling, dynamically pruning low-contribution Gaussians to reduce rendering overhead; and (3) a view-dependent scene compression strategy that significantly lowers storage requirements. Experiments demonstrate that our approach achieves a 2.6× rendering speedup on mainstream mobile GPUs, reduces model size by 32.3%, and maintains superior rendering fidelity compared to state-of-the-art methods. To the best of our knowledge, this is the first work to enable high-quality real-time 3DGS rendering on mobile platforms.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Hardware-aware ML
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

3D Gaussian Splatting (3DGS) has become a crucial rendering technique for many real-time applications. However, the limited hardware resources on today's mobile platforms hinder these applications, as they struggle to achieve real-time performance. In this paper, we propose SeeLe, a general framework designed to accelerate the 3DGS pipeline for resource-constrained mobile devices. Specifically, we propose two GPU-oriented techniques: hybrid preprocessing and contribution-aware rasterization. Hybrid preprocessing alleviates the GPU compute and memory pressure by reducing the number of irrelevant Gaussians during rendering. The key is to combine our view-dependent scene representation with online filtering. Meanwhile, contribution-aware rasterization improves the GPU utilization at the rasterization stage by prioritizing Gaussians with high contributions while reducing computations for those with low contributions. Both techniques can be seamlessly integrated into existing 3DGS pipelines with minimal fine-tuning. Collectively, our framework achieves 2.6$ imes$ speedup and 32.3% model reduction while achieving superior rendering quality compared to existing methods.

Problem

• Accelerates 3D Gaussian Splatting for mobile devices
• Reduces GPU compute and memory pressure
• Improves rendering quality and performance

Innovation

• Hybrid preprocessing reduces GPU compute pressure.
• Contribution-aware rasterization improves GPU utilization.
• Seamless integration into existing 3DGS pipelines.