ADC-GS: Anchor-Driven Deformable and Compressed Gaussian Splatting for Dynamic Scene Reconstruction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, He, Yang, Qi, Liu, Mufan, Xu, Yiling, Li, Zhu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912373233680384
author Huang, He
Yang, Qi
Liu, Mufan
Xu, Yiling
Li, Zhu
author_facet Huang, He
Yang, Qi
Liu, Mufan
Xu, Yiling
Li, Zhu
contents Existing 4D Gaussian Splatting methods rely on per-Gaussian deformation from a canonical space to target frames, which overlooks redundancy among adjacent Gaussian primitives and results in suboptimal performance. To address this limitation, we propose Anchor-Driven Deformable and Compressed Gaussian Splatting (ADC-GS), a compact and efficient representation for dynamic scene reconstruction. Specifically, ADC-GS organizes Gaussian primitives into an anchor-based structure within the canonical space, enhanced by a temporal significance-based anchor refinement strategy. To reduce deformation redundancy, ADC-GS introduces a hierarchical coarse-to-fine pipeline that captures motions at varying granularities. Moreover, a rate-distortion optimization is adopted to achieve an optimal balance between bitrate consumption and representation fidelity. Experimental results demonstrate that ADC-GS outperforms the per-Gaussian deformation approaches in rendering speed by 300%-800% while achieving state-of-the-art storage efficiency without compromising rendering quality. The code is released at https://github.com/H-Huang774/ADC-GS.git.
format Preprint
id arxiv_https___arxiv_org_abs_2505_08196
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ADC-GS: Anchor-Driven Deformable and Compressed Gaussian Splatting for Dynamic Scene Reconstruction
Huang, He
Yang, Qi
Liu, Mufan
Xu, Yiling
Li, Zhu
Computer Vision and Pattern Recognition
Existing 4D Gaussian Splatting methods rely on per-Gaussian deformation from a canonical space to target frames, which overlooks redundancy among adjacent Gaussian primitives and results in suboptimal performance. To address this limitation, we propose Anchor-Driven Deformable and Compressed Gaussian Splatting (ADC-GS), a compact and efficient representation for dynamic scene reconstruction. Specifically, ADC-GS organizes Gaussian primitives into an anchor-based structure within the canonical space, enhanced by a temporal significance-based anchor refinement strategy. To reduce deformation redundancy, ADC-GS introduces a hierarchical coarse-to-fine pipeline that captures motions at varying granularities. Moreover, a rate-distortion optimization is adopted to achieve an optimal balance between bitrate consumption and representation fidelity. Experimental results demonstrate that ADC-GS outperforms the per-Gaussian deformation approaches in rendering speed by 300%-800% while achieving state-of-the-art storage efficiency without compromising rendering quality. The code is released at https://github.com/H-Huang774/ADC-GS.git.
title ADC-GS: Anchor-Driven Deformable and Compressed Gaussian Splatting for Dynamic Scene Reconstruction
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.08196