Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2508.02129 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910201055019008 |
|---|---|
| author | Xiao, Yuru Lin, Zihan Lu, Chao Zhai, Deming Jiang, Kui Zhao, Wenbo Zhang, Wei Jiang, Junjun Wang, Huanran Liu, Xianming |
| author_facet | Xiao, Yuru Lin, Zihan Lu, Chao Zhai, Deming Jiang, Kui Zhao, Wenbo Zhang, Wei Jiang, Junjun Wang, Huanran Liu, Xianming |
| contents | Dynamic urban scene modeling is a rapidly evolving area with broad applications. While current approaches leveraging neural radiance fields or Gaussian Splatting have achieved fine-grained reconstruction and high-fidelity novel view synthesis, they still face significant limitations. These often stem from a dependence on pre-calibrated object tracks or difficulties in accurately modeling fast-moving objects from undersampled capture, particularly due to challenges in handling temporal discontinuities. To overcome these issues, we propose a novel video diffusion-enhanced 4D Gaussian Splatting framework. Our key insight is to distill robust, temporally consistent priors from a test-time adapted video diffusion model. To ensure precise pose alignment and effective integration of this denoised content, we introduce two core innovations: a joint timestamp optimization strategy that refines interpolated frame poses, and an uncertainty distillation method that adaptively extracts target content while preserving well-reconstructed regions. Extensive experiments demonstrate that our method significantly enhances dynamic modeling, especially for fast-moving objects, achieving an approximate PSNR gain of 2 dB for novel view synthesis over baseline approaches. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_02129 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | VDEGaussian: Video Diffusion Enhanced 4D Gaussian Splatting for Dynamic Urban Scenes Modeling Xiao, Yuru Lin, Zihan Lu, Chao Zhai, Deming Jiang, Kui Zhao, Wenbo Zhang, Wei Jiang, Junjun Wang, Huanran Liu, Xianming Computer Vision and Pattern Recognition Dynamic urban scene modeling is a rapidly evolving area with broad applications. While current approaches leveraging neural radiance fields or Gaussian Splatting have achieved fine-grained reconstruction and high-fidelity novel view synthesis, they still face significant limitations. These often stem from a dependence on pre-calibrated object tracks or difficulties in accurately modeling fast-moving objects from undersampled capture, particularly due to challenges in handling temporal discontinuities. To overcome these issues, we propose a novel video diffusion-enhanced 4D Gaussian Splatting framework. Our key insight is to distill robust, temporally consistent priors from a test-time adapted video diffusion model. To ensure precise pose alignment and effective integration of this denoised content, we introduce two core innovations: a joint timestamp optimization strategy that refines interpolated frame poses, and an uncertainty distillation method that adaptively extracts target content while preserving well-reconstructed regions. Extensive experiments demonstrate that our method significantly enhances dynamic modeling, especially for fast-moving objects, achieving an approximate PSNR gain of 2 dB for novel view synthesis over baseline approaches. |
| title | VDEGaussian: Video Diffusion Enhanced 4D Gaussian Splatting for Dynamic Urban Scenes Modeling |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.02129 |