HAC++: Towards 100X Compression of 3D Gaussian Splatting

3D Gaussian Splatting (3DGS) has emerged as a promising representation for novel view synthesis, boosting rapid rendering speed with high fidelity. However, the substantial Gaussians and their associated attributes necessitate effective compression techniques. Nevertheless, the sparse and unorganize...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on pattern analysis and machine intelligence Vol. 47; no. 11; pp. 10210 - 10226
Main Authors: Chen, Yihang, Wu, Qianyi, Lin, Weiyao, Harandi, Mehrtash, Cai, Jianfei
Format: Journal Article
Language:English
Published: United States IEEE 01.11.2025
Subjects:
3D gaussian splatting (3DGS)
Adaptation models
compression
context model
Context modeling
Entropy
Entropy coding
Mutual information
Neural radiance field
Redundancy
Rendering (computer graphics)
Three-dimensional displays
Training
ISSN:0162-8828, 1939-3539, 2160-9292, 1939-3539
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:3D Gaussian Splatting (3DGS) has emerged as a promising representation for novel view synthesis, boosting rapid rendering speed with high fidelity. However, the substantial Gaussians and their associated attributes necessitate effective compression techniques. Nevertheless, the sparse and unorganized nature of the point cloud of Gaussians (or anchors in our paper) presents challenges for compression. In this paper, we propose HAC++, which explicitly minimizes the representation's entropy during optimization, enabling efficient arithmetic coding after training for compressed storage. Specifically, to reduce entropy, HAC++ leverages the relationships between unorganized anchors and a structured hash grid, utilizing their mutual information for context modeling. Additionally, HAC++ captures intra-anchor contextual relationships to further enhance compression performance. To facilitate entropy coding, we utilize Gaussian distributions to precisely estimate the probability of each quantized attribute, where an adaptive quantization module is proposed to enable high-precision quantization of these attributes for improved fidelity restoration. Moreover, we incorporate an adaptive masking strategy to eliminate non-effective Gaussians and anchors. Overall, HAC++ achieves a remarkable size reduction of over <inline-formula><tex-math notation="LaTeX">100\times</tex-math> <mml:math><mml:mrow><mml:mn>100</mml:mn><mml:mo>×</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="lin-ieq1-3594066.gif"/> </inline-formula> compared to vanilla 3DGS when averaged on all datasets, while simultaneously improving fidelity. It also delivers more than <inline-formula><tex-math notation="LaTeX">20\times</tex-math> <mml:math><mml:mrow><mml:mn>20</mml:mn><mml:mo>×</mml:mo></mml:mrow></mml:math><inline-graphic xlink:href="lin-ieq2-3594066.gif"/> </inline-formula> size reduction compared to Scaffold-GS.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0162-8828
1939-3539
2160-9292
1939-3539
DOI:10.1109/TPAMI.2025.3594066