DeGS: A Scalable 3DGS Architecture via Decoupled Workload Parsing and Reorganization
2026-08-03 • Hardware Architecture
Hardware ArchitectureArtificial IntelligenceComputer Vision and Pattern Recognition
AI summaryⓘ
The authors studied a method called 3D Gaussian Splatting used to create new 3D views quickly but found that existing designs do not speed up much when adding more processing units. They discovered this is because the current way of combining steps in rendering causes inefficiencies and underused hardware. Their solution, DeGS, separates these steps into distinct stages to better organize the workload, making hardware use much more efficient. Tested on real hardware, DeGS shows several times faster processing and better energy use than previous designs, especially when using many processors.
3D Gaussian Splattingnovel view synthesisprocessing elements (PEs)dataflowalpha blendingtransmittance checkinghardware acceleratorparallel executionenergy efficiencyworkload reorganization
Authors
Minnan Pei, Gang Li, Zeyu Zhu, Siting Wang, Junwen Si, Zhuoran Song, Yu Feng, Fangxin Liu, Xiaoyao Liang, Jian Cheng
Abstract
3D Gaussian Splatting (3DGS) has emerged as a leading technique for real-time novel view synthesis, yet existing 3DGS accelerators suffer from poor architectural scalability: increasing the number of PEs leads to marginal performance improvement during rendering. We identify that the root cause is the tightly coupled ``checking-while-blending'' dataflow, which exacerbates PE underutilization caused by spatial redundancy from irregular Gaussian coverage and temporal redundancy from asynchronous pixel-wise termination under parallel execution. To address this issue, we propose DeGS, a scalable architecture for efficient 3DGS inference. To systematically eliminate the redundancies inherent in rendering, DeGS exploits a decoupled dataflow, restructuring the coupled $α$-checking, transmittance checking, and $α$-blending of the standard rendering process into consecutive workload parsing, reorganization, and blending stages. This allows the fragmented, length-variable, and temporal-dependent workloads to be reorganized into compact, conflict-free, and dense workloads prior to blending, thereby significantly improving PE utilization during parallel blending. Implemented in 28 nm technology, DeGS achieves 2.36$\times$--7.25$\times$ throughput, 1.82$\times$--6.02$\times$ end-to-end speedup, and 1.59$\times$--4.42$\times$ energy efficiency over state-of-the-art 3DGS accelerators (GSCore, GBU, GCC) across diverse scenes and resolutions (720p to 8K). Moreover, scaling from 16 to 1024 PEs, DeGS maintains over 80\% PE utilization at high resolutions, significantly outperforming existing accelerators.