Saved in:
Bibliographic Details
Main Authors: Yin, Zeyuan, Liu, Xiaoming
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.16642
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912721423826944
author Yin, Zeyuan
Liu, Xiaoming
author_facet Yin, Zeyuan
Liu, Xiaoming
contents Recent advances in 3D Gaussian diffusion models suffer from time-intensive denoising and post-denoising processing due to the massive number of Gaussian primitives, resulting in slow generation and limited scalability along sampling trajectories. To improve the efficiency of 3D diffusion models, we propose $\textbf{TRIM}$ ($\textbf{T}$rajectory $\textbf{R}$eduction and $\textbf{I}$nstance $\textbf{M}$ask denoising), a post-training approach that incorporates both temporal and spatial trimming strategies, to accelerate inference without compromising output quality while supporting the inference-time scaling for Gaussian diffusion models. Instead of scaling denoising trajectories in a costly end-to-end manner, we develop a lightweight selector model to evaluate latent Gaussian primitives derived from multiple sampled noises, enabling early trajectory reduction by selecting candidates with high-quality potential. Furthermore, we introduce instance mask denoising to prune learnable Gaussian primitives by filtering out redundant background regions, reducing inference computation at each denoising step. Extensive experiments and analysis demonstrate that TRIM significantly improves both the efficiency and quality of 3D generation. Source code is available at $\href{https://github.com/zeyuanyin/TRIM}{link}$.
format Preprint
id arxiv_https___arxiv_org_abs_2511_16642
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TRIM: Scalable 3D Gaussian Diffusion Inference with Temporal and Spatial Trimming
Yin, Zeyuan
Liu, Xiaoming
Computer Vision and Pattern Recognition
Recent advances in 3D Gaussian diffusion models suffer from time-intensive denoising and post-denoising processing due to the massive number of Gaussian primitives, resulting in slow generation and limited scalability along sampling trajectories. To improve the efficiency of 3D diffusion models, we propose $\textbf{TRIM}$ ($\textbf{T}$rajectory $\textbf{R}$eduction and $\textbf{I}$nstance $\textbf{M}$ask denoising), a post-training approach that incorporates both temporal and spatial trimming strategies, to accelerate inference without compromising output quality while supporting the inference-time scaling for Gaussian diffusion models. Instead of scaling denoising trajectories in a costly end-to-end manner, we develop a lightweight selector model to evaluate latent Gaussian primitives derived from multiple sampled noises, enabling early trajectory reduction by selecting candidates with high-quality potential. Furthermore, we introduce instance mask denoising to prune learnable Gaussian primitives by filtering out redundant background regions, reducing inference computation at each denoising step. Extensive experiments and analysis demonstrate that TRIM significantly improves both the efficiency and quality of 3D generation. Source code is available at $\href{https://github.com/zeyuanyin/TRIM}{link}$.
title TRIM: Scalable 3D Gaussian Diffusion Inference with Temporal and Spatial Trimming
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.16642