Temporal Gaussian Hierarchy

Creator
Creator
Seonglae ChoSeonglae Cho
Created
Created
2025 Jun 8 16:34
Editor
Edited
Edited
2025 Jun 8 16:37

TGH

By organizing 4D Gaussians into a hierarchical structure, memory and computation are optimized by loading only the primitives needed for specific timestamps.
Using temporally hierarchical Gaussian primitives, long (several minutes) volumetric videos can be efficiently modeled. Each level handles scene regions that change at different rates, reducing unnecessary redundancy, and only the primitives needed for specific timestamps are loaded to GPU, keeping memory usage almost constant. Compact Appearance Model: View-dependent effects are represented while significantly reducing model size by combining base colors with sparse SH coefficients. Using gradient thresholds, SH is optimized only for Gaussians that actually need it.
 
 
 
 
Representing Long Volumetric Video with Temporal Gaussian Hierarchy
This paper aims to address the challenge of reconstructing long volumetric videos from multi-view RGB videos. Recent dynamic view synthesis methods leverage powerful 4D representations, like feature grids or point cloud sequences, to achieve high-quality rendering results. However, they are typically limited to short (1∼similar-to\sim∼2s) video clips and often suffer from large memory footprints when dealing with longer videos. To solve this issue, we propose a novel 4D representation, named Temporal Gaussian Hierarchy, to compactly model long volumetric videos. Our key observation is that there are generally various degrees of temporal redundancy in dynamic scenes, which consist of areas changing at different speeds. Motivated by this, our approach builds a multi-level hierarchy of 4D Gaussian primitives, where each level separately describes scene regions with different degrees of content change, and adaptively shares Gaussian primitives to represent unchanged scene content over different temporal segments, thus effectively reducing the number of Gaussian primitives. In addition, the tree-like structure of the Gaussian hierarchy allows us to efficiently represent the scene at a particular moment with a subset of Gaussian primitives, leading to nearly constant GPU memory usage during the training or rendering regardless of the video length. Moreover, we design a Compact Appearance Model that mixes diffuse and view-dependent Gaussians to further minimize the model size while maintaining the rendering quality. We also develop a rasterization pipeline of Gaussian primitives based on the hardware-accelerated technique to improve rendering speed. Extensive experimental results demonstrate the superiority of our method over alternative methods in terms of training cost, rendering speed, and storage usage. To our knowledge, this work is the first approach capable of efficiently handling minutes of volumetric video data while maintaining state-of-the-art rendering quality.
 
 
 

Backlinks

4DVLab

Recommendations