Salvato in:
Dettagli Bibliografici
Autori principali: Tseng, Yu-Jen, Kao, Chia-Hao, Chen, Jing-Zhong, Gnutti, Alessandro, Lo, Shao-Yuan, Lin, Yen-Yu, Peng, Wen-Hsiao
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2601.12814
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912832042303488
author Tseng, Yu-Jen
Kao, Chia-Hao
Chen, Jing-Zhong
Gnutti, Alessandro
Lo, Shao-Yuan
Lin, Yen-Yu
Peng, Wen-Hsiao
author_facet Tseng, Yu-Jen
Kao, Chia-Hao
Chen, Jing-Zhong
Gnutti, Alessandro
Lo, Shao-Yuan
Lin, Yen-Yu
Peng, Wen-Hsiao
contents We present the first unified framework for rate-distortion-optimized compression and segmentation of 3D Gaussian Splatting (3DGS). While 3DGS has proven effective for both real-time rendering and semantic scene understanding, prior works have largely treated these tasks independently, leaving their joint consideration unexplored. Inspired by recent advances in rate-distortion-optimized 3DGS compression, this work integrates semantic learning into the compression pipeline to support decoder-side applications--such as scene editing and manipulation--that extend beyond traditional scene reconstruction and view synthesis. Our scheme features a lightweight implicit neural representation-based hyperprior, enabling efficient entropy coding of both color and semantic attributes while avoiding costly grid-based hyperprior as seen in many prior works. To facilitate compression and segmentation, we further develop compression-guided segmentation learning, consisting of quantization-aware training to enhance feature separability and a quality-aware weighting mechanism to suppress unreliable Gaussian primitives. Extensive experiments on the LERF and 3D-OVS datasets demonstrate that our approach significantly reduces transmission cost while preserving high rendering quality and strong segmentation performance.
format Preprint
id arxiv_https___arxiv_org_abs_2601_12814
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle CSGaussian: Progressive Rate-Distortion Compression and Segmentation for 3D Gaussian Splatting
Tseng, Yu-Jen
Kao, Chia-Hao
Chen, Jing-Zhong
Gnutti, Alessandro
Lo, Shao-Yuan
Lin, Yen-Yu
Peng, Wen-Hsiao
Computer Vision and Pattern Recognition
We present the first unified framework for rate-distortion-optimized compression and segmentation of 3D Gaussian Splatting (3DGS). While 3DGS has proven effective for both real-time rendering and semantic scene understanding, prior works have largely treated these tasks independently, leaving their joint consideration unexplored. Inspired by recent advances in rate-distortion-optimized 3DGS compression, this work integrates semantic learning into the compression pipeline to support decoder-side applications--such as scene editing and manipulation--that extend beyond traditional scene reconstruction and view synthesis. Our scheme features a lightweight implicit neural representation-based hyperprior, enabling efficient entropy coding of both color and semantic attributes while avoiding costly grid-based hyperprior as seen in many prior works. To facilitate compression and segmentation, we further develop compression-guided segmentation learning, consisting of quantization-aware training to enhance feature separability and a quality-aware weighting mechanism to suppress unreliable Gaussian primitives. Extensive experiments on the LERF and 3D-OVS datasets demonstrate that our approach significantly reduces transmission cost while preserving high rendering quality and strong segmentation performance.
title CSGaussian: Progressive Rate-Distortion Compression and Segmentation for 3D Gaussian Splatting
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.12814