ATGS: Anchored Temporal Gaussian Splatting for Long Volumetric Video Representation
Open Access
- 3 July 2026
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 45 (4) , 1-13
Abstract
Volumetric video enables immersive free viewpoint rendering of dynamic real world scenes, yet existing methods struggle with long sequences and complex motions, often leading to temporal instability and visual artifacts. To address these challenges, we propose ATGS, a Gaussian splatting based framework for volumetric video reconstruction. Our key insight is that explicitly tracking long term complex motion with individual Gaussian primitives is inherently unstable. Instead, we organize Gaussians around time conditioned anchors that localize their spatial and temporal support, thereby reducing long range motion complexity. We further introduce a temporal windowing strategy to activate only anchors relevant to the queried time, which improves scalability and temporal coherence. In addition, to ensure spatial and temporal stability, we design a compact set of multi level anchor features that encode global features, local spatial features, and local temporal features, jointly constraining Gaussian generation. Extensive experiments demonstrate that ATGS consistently outperforms prior methods on long sequence volumetric videos with complex motions. Project page: https://github.com/WuJH2001/ATGS.Keywords
All Related Versions
Funding Information
- Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China (JYB2025XDXM413)
- Guangdong Provincial Key Laboratory of Ultra High Definition Immersive Media Technology (2024B1212010006)
- Outstanding Talents Training Fund in Shenzhen, Shenzhen Science and Technology Program (SYSPG20241211173440004)
- Outstanding Talents Training Fund in Shenzhen, Shenzhen Science and Technology Program (KJZD20230923114707016)
This publication has 33 references indexed in Scilit:
- Dynamic 3D Gaussians: Tracking by Persistent Dynamic View Synthesis2024
- HexPlane: A Fast Representation for Dynamic Scenes2023
- TensoRF: Tensorial Radiance Fields2022
- Neural 3D Video Synthesis from Multi-view Video2022
- Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstruction2022
- Plenoxels: Radiance Fields without Neural Networks2022
- D-NeRF: Neural Radiance Fields for Dynamic Scenes2021
- Neural Scene Flow Fields for Space-Time View Synthesis of Dynamic Scenes2021
- DynamicFusion: Reconstruction and tracking of non-rigid scenes in real-time2015
- Dense correspondence finding for parametrization-free animation reconstruction from video2008