2026/04/11 by Fang-Chi Chang, Tian‐Sheuan Chang, Tian-Sheuan Chang · 1 voice
Computer Science · Engineering · Physics and Astronomy · #Advanced Optical Sensing Technologies #Photorefractive and Nonlinear Optics #Radiation Detection and Scintillator Technologies #cs.AR #cs.GR #eess.IV
paper · pdf · doi:10.1109/tvcg.2026.3683714
arxiv published 2026/04/11 · arxiv updated 2026/04/11 · openalex publication_date 2026/04/16 · openalex created_date 2026/04/17 · openalex updated_date 2026/07/29
Rendering large-scale, unbounded scenes on AR/VR-class devices is constrained by the computation, bandwidth, and storage cost of 3D Gaussian Splatting (3DGS). We propose a low-power, low-cost 3DGS hardware accelerator that renders full-HD images in real time, together with a hardware-friendly compression pipeline that combines iterative Gaussian pruning and fine-tuning, progressive spherical harmonics (SH) degree reduction, and vector quantization of all SH coefficients and colors. The scheme achieves a 51.6×51.6× model-size reduction with a 0.743 dB PSNR loss. The accelerator uses a frame-level pipeline that integrates point-based culling and projection with tile-based sorting and rasterization, skips zero-Jacobian matrix multiplications (reducing processing elements by 63% and computation by 53%), and adopts comparison-free tile-based sorting with deterministic latency. Implemented in a TSMC 28-nm process at 800MHz, the design occupies 0.66 mm20.66mm2 with 1.1438 M gates and 120 kB SRAM, consumes 0.219 W, and delivers 1219 Mpixels/J at 267.5 Mpixels/s, enabling 1080p at 129 FPS. Overall, it is 5.98×5.98× smaller in area, 5.94×5.94× higher throughput, and delivers 7.5×7.5× higher energy efficiency than prior 3DGS accelerators.