AI Paper Atlas

Research map / 3D vision / Gaussian splatting

Residual-Autoregressive Context for 3D Gaussian Splatting Compression

Wenqing Wang, Huimin Zeng, Yun Fu

NeurIPS 2026 · 3D vision › Gaussian splatting · approach: Compact and efficient Gaussian splatting

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes RacerGS, a compact 3D Gaussian Splatting compression framework that reduces storage while preserving rendering fidelity.

Problem

The large parameter size of 3D Gaussian Splatting scenes is a bottleneck for storage and transmission.

Approach

Compresses 3DGS scenes with a residual-autoregressive context model intended to yield compact and spatially coherent features; the abstract is truncated, so further details are unavailable.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add adaptive residual quantization (encode residuals around context-predicted means) (ablation)RacerGS · Mip-NeRF360 (high rate)size (MB): 18.24 -> 13.61Improved
add hash projection MLP before the entropy model (ablation)RacerGS · Mip-NeRF360 (high rate)PSNR (dB) / size (MB): 27.62 -> 27.84 dB; 15.21 -> 13.61 MBImproved
use MLP projection instead of a single linear projection (ablation)RacerGS hash projection · Mip-NeRF360 (high rate)PSNR (dB) / size (MB): 27.76 -> 27.84 dB; 14.28 -> 13.61 MBImproved
vary context dimension D^c from 12 to 48 (ablation)RacerGS · Mip-NeRF360 (high rate)PSNR (dB) / size (MB): size at most 0.35 MB, PSNR at most 0.04 dBNo clear effect
replace decoded-chunk concatenation with GRU context model (ablation)RacerGS · Mip-NeRF360 (high rate)size (MB): 16.87 -> 13.61Improved
add FiLM modulation to the GRU context model (ablation)RacerGS GRU · Mip-NeRF360 (high rate)size (MB): 17.32 -> 13.61Improved
GRU+FiLM instead of capacity-matched concatenation (ablation)RacerGS context model · Mip-NeRF360 (high rate)size (MB): 15.24 -> 13.61Improved
add an intra-anchor context model (ablation)RacerGS · Mip-NeRF360 (high rate)size (MB): 18.20 -> 13.61Improved
GRU+FiLM versus concatenation for later chunks (ablation)RacerGS · Mip-NeRF360 (high rate)bits per anchor, chunk 5: 80.1 -> 39.0Improved
residual coding of anchor features instead of attribute coding (ablation)RacerGS · Mip-NeRF360 (high rate)coded bits per element: 8.44 -> 5.38Improved
use RacerGS instead of HAC++ at high rateanchor-based 3DGS compression · Mip-NeRF360 (high rate)size (MB): 18.48 -> 13.61 MB, PSNR 27.82 -> 27.84Improved
RacerGS versus HAC++ in training timeRacerGS · Mip-NeRF360training time (s): 2447 -> 2850Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy