ContextGS: Compact 3D Gaussian Splatting with Anchor Level Context Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Yufei, Li, Zhihao, Guo, Lanqing, Yang, Wenhan, Kot, Alex C., Wen, Bihan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929367880302592
author Wang, Yufei
Li, Zhihao
Guo, Lanqing
Yang, Wenhan
Kot, Alex C.
Wen, Bihan
author_facet Wang, Yufei
Li, Zhihao
Guo, Lanqing
Yang, Wenhan
Kot, Alex C.
Wen, Bihan
contents Recently, 3D Gaussian Splatting (3DGS) has become a promising framework for novel view synthesis, offering fast rendering speeds and high fidelity. However, the large number of Gaussians and their associated attributes require effective compression techniques. Existing methods primarily compress neural Gaussians individually and independently, i.e., coding all the neural Gaussians at the same time, with little design for their interactions and spatial dependence. Inspired by the effectiveness of the context model in image compression, we propose the first autoregressive model at the anchor level for 3DGS compression in this work. We divide anchors into different levels and the anchors that are not coded yet can be predicted based on the already coded ones in all the coarser levels, leading to more accurate modeling and higher coding efficiency. To further improve the efficiency of entropy coding, e.g., to code the coarsest level with no already coded anchors, we propose to introduce a low-dimensional quantized feature as the hyperprior for each anchor, which can be effectively compressed. Our work pioneers the context model in the anchor level for 3DGS representation, yielding an impressive size reduction of over 100 times compared to vanilla 3DGS and 15 times compared to the most recent state-of-the-art work Scaffold-GS, while achieving comparable or even higher rendering quality.
format Preprint
id arxiv_https___arxiv_org_abs_2405_20721
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ContextGS: Compact 3D Gaussian Splatting with Anchor Level Context Model
Wang, Yufei
Li, Zhihao
Guo, Lanqing
Yang, Wenhan
Kot, Alex C.
Wen, Bihan
Computer Vision and Pattern Recognition
Artificial Intelligence
Recently, 3D Gaussian Splatting (3DGS) has become a promising framework for novel view synthesis, offering fast rendering speeds and high fidelity. However, the large number of Gaussians and their associated attributes require effective compression techniques. Existing methods primarily compress neural Gaussians individually and independently, i.e., coding all the neural Gaussians at the same time, with little design for their interactions and spatial dependence. Inspired by the effectiveness of the context model in image compression, we propose the first autoregressive model at the anchor level for 3DGS compression in this work. We divide anchors into different levels and the anchors that are not coded yet can be predicted based on the already coded ones in all the coarser levels, leading to more accurate modeling and higher coding efficiency. To further improve the efficiency of entropy coding, e.g., to code the coarsest level with no already coded anchors, we propose to introduce a low-dimensional quantized feature as the hyperprior for each anchor, which can be effectively compressed. Our work pioneers the context model in the anchor level for 3DGS representation, yielding an impressive size reduction of over 100 times compared to vanilla 3DGS and 15 times compared to the most recent state-of-the-art work Scaffold-GS, while achieving comparable or even higher rendering quality.
title ContextGS: Compact 3D Gaussian Splatting with Anchor Level Context Model
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2405.20721