Splat4D: Diffusion-Enhanced 4D Gaussian Splatting for Temporally and Spatially Consistent Content Creation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yin, Minghao, Cao, Yukang, Peng, Songyou, Han, Kai
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912531369426944
author Yin, Minghao
Cao, Yukang
Peng, Songyou
Han, Kai
author_facet Yin, Minghao
Cao, Yukang
Peng, Songyou
Han, Kai
contents Generating high-quality 4D content from monocular videos for applications such as digital humans and AR/VR poses challenges in ensuring temporal and spatial consistency, preserving intricate details, and incorporating user guidance effectively. To overcome these challenges, we introduce Splat4D, a novel framework enabling high-fidelity 4D content generation from a monocular video. Splat4D achieves superior performance while maintaining faithful spatial-temporal coherence by leveraging multi-view rendering, inconsistency identification, a video diffusion model, and an asymmetric U-Net for refinement. Through extensive evaluations on public benchmarks, Splat4D consistently demonstrates state-of-the-art performance across various metrics, underscoring the efficacy of our approach. Additionally, the versatility of Splat4D is validated in various applications such as text/image conditioned 4D generation, 4D human generation, and text-guided content editing, producing coherent outcomes following user instructions.
format Preprint
id arxiv_https___arxiv_org_abs_2508_07557
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Splat4D: Diffusion-Enhanced 4D Gaussian Splatting for Temporally and Spatially Consistent Content Creation
Yin, Minghao
Cao, Yukang
Peng, Songyou
Han, Kai
Computer Vision and Pattern Recognition
Generating high-quality 4D content from monocular videos for applications such as digital humans and AR/VR poses challenges in ensuring temporal and spatial consistency, preserving intricate details, and incorporating user guidance effectively. To overcome these challenges, we introduce Splat4D, a novel framework enabling high-fidelity 4D content generation from a monocular video. Splat4D achieves superior performance while maintaining faithful spatial-temporal coherence by leveraging multi-view rendering, inconsistency identification, a video diffusion model, and an asymmetric U-Net for refinement. Through extensive evaluations on public benchmarks, Splat4D consistently demonstrates state-of-the-art performance across various metrics, underscoring the efficacy of our approach. Additionally, the versatility of Splat4D is validated in various applications such as text/image conditioned 4D generation, 4D human generation, and text-guided content editing, producing coherent outcomes following user instructions.
title Splat4D: Diffusion-Enhanced 4D Gaussian Splatting for Temporally and Spatially Consistent Content Creation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.07557