HybridGS: High-Efficiency Gaussian Splatting Data Compression using Dual-Channel Sparse Representation and Point Cloud Encoder

Qi Yang, Le Yang, Geert Van Der Auwera, Zhu Li
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:71487-71507, 2025.

Abstract

Most existing 3D Gaussian Splatting (3DGS) compression schemes focus on producing compact 3DGS representation via implicit data embedding. They have long encoding and decoding times and highly customized data format, making it difficult for widespread deployment. This paper presents a new 3DGS compression framework called HybridGS, which takes advantage of both compact generation and standardized point cloud data encoding. HybridGS first generates compact and explicit 3DGS data. A dual-channel sparse representation is introduced to supervise the primitive position and feature bit depth. It then utilizes a canonical point cloud encoder to carry out further data compression and form standard output bitstreams. A simple and effective rate control scheme is proposed to pivot the interpretable data compression scheme. HybridGS does not include any modules aimed at improving 3DGS quality during generation. But experiment results show that it still provides comparable reconstruction performance against state-of-the-art methods, with evidently faster encoding and decoding speed. The code is publicly available at https://github.com/Qi-Yangsjtu/HybridGS .

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-yang25ar, title = {{H}ybrid{GS}: High-Efficiency {G}aussian Splatting Data Compression using Dual-Channel Sparse Representation and Point Cloud Encoder}, author = {Yang, Qi and Yang, Le and Van Der Auwera, Geert and Li, Zhu}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {71487--71507}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/yang25ar/yang25ar.pdf}, url = {https://proceedings.mlr.press/v267/yang25ar.html}, abstract = {Most existing 3D Gaussian Splatting (3DGS) compression schemes focus on producing compact 3DGS representation via implicit data embedding. They have long encoding and decoding times and highly customized data format, making it difficult for widespread deployment. This paper presents a new 3DGS compression framework called HybridGS, which takes advantage of both compact generation and standardized point cloud data encoding. HybridGS first generates compact and explicit 3DGS data. A dual-channel sparse representation is introduced to supervise the primitive position and feature bit depth. It then utilizes a canonical point cloud encoder to carry out further data compression and form standard output bitstreams. A simple and effective rate control scheme is proposed to pivot the interpretable data compression scheme. HybridGS does not include any modules aimed at improving 3DGS quality during generation. But experiment results show that it still provides comparable reconstruction performance against state-of-the-art methods, with evidently faster encoding and decoding speed. The code is publicly available at https://github.com/Qi-Yangsjtu/HybridGS .} }
Endnote
%0 Conference Paper %T HybridGS: High-Efficiency Gaussian Splatting Data Compression using Dual-Channel Sparse Representation and Point Cloud Encoder %A Qi Yang %A Le Yang %A Geert Van Der Auwera %A Zhu Li %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-yang25ar %I PMLR %P 71487--71507 %U https://proceedings.mlr.press/v267/yang25ar.html %V 267 %X Most existing 3D Gaussian Splatting (3DGS) compression schemes focus on producing compact 3DGS representation via implicit data embedding. They have long encoding and decoding times and highly customized data format, making it difficult for widespread deployment. This paper presents a new 3DGS compression framework called HybridGS, which takes advantage of both compact generation and standardized point cloud data encoding. HybridGS first generates compact and explicit 3DGS data. A dual-channel sparse representation is introduced to supervise the primitive position and feature bit depth. It then utilizes a canonical point cloud encoder to carry out further data compression and form standard output bitstreams. A simple and effective rate control scheme is proposed to pivot the interpretable data compression scheme. HybridGS does not include any modules aimed at improving 3DGS quality during generation. But experiment results show that it still provides comparable reconstruction performance against state-of-the-art methods, with evidently faster encoding and decoding speed. The code is publicly available at https://github.com/Qi-Yangsjtu/HybridGS .
APA
Yang, Q., Yang, L., Van Der Auwera, G. & Li, Z.. (2025). HybridGS: High-Efficiency Gaussian Splatting Data Compression using Dual-Channel Sparse Representation and Point Cloud Encoder. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:71487-71507 Available from https://proceedings.mlr.press/v267/yang25ar.html.

Related Material