Abstract
3D Gaussian Splatting (3DGS) has gained popularity for its efficiency and sparse Gaussian-based representation. However, 3DGS struggles to meet the real-time requirement of 90 frames per second (FPS) on resource-constrained mobile devices, achieving only 2 to 9 FPS.Existing accelerators focus on compute efficiency but overlook memory efficiency, leading to redundant DRAM traffic. We introduce STREAMINGGS, a fully streaming 3DGS algorithm-architecture co-design that achieves fine-grained pipelining and reduces DRAM traffic by transforming from a tile-centric rendering to a memory-centric rendering. Results show that our design achieves up to 45.7 $\times$ speedup and 62.9 $\times$ energy savings over mobile Ampere GPUs.
Related Papers
SGLoc: Semantic Localization System for Camera Pose Estimation from 3D Gaussian Splatting Representation2025-07-163DGAA: Realistic and Robust 3D Gaussian-based Adversarial Attack for Autonomous Driving2025-07-14LighthouseGS: Indoor Structure-aware 3D Gaussian Splatting for Panorama-Style Mobile Captures2025-07-08Reflections Unlock: Geometry-Aware Reflection Disentanglement in 3D Gaussian Splatting for Photorealistic Scenes Rendering2025-07-083DGS_LSR:Large_Scale Relocation for Autonomous Driving Based on 3D Gaussian Splatting2025-07-08Outdoor Monocular SLAM with Global Scale-Consistent 3D Gaussian Pointmaps2025-07-043D Gaussian Splatting Driven Multi-View Robust Physical Adversarial Camouflage Generation2025-07-02TVG-SLAM: Robust Gaussian Splatting SLAM with Tri-view Geometric Constraints2025-06-29