4DVGGT-D: 4D Visual Geometry Transformer with Improved Dynamic Depth Estimation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910212167827456 |
|---|---|
| author | Zang, Ying Liu, Xuanyi Han, Yidong Ji, Deyi Ding, Chaotao Hu, Yuanqi Zhu, Qi Li, Xuanfu Ma, Jin Sun, Lingyun Chen, Tianrun Zhu, Lanyun |
| author_facet | Zang, Ying Liu, Xuanyi Han, Yidong Ji, Deyi Ding, Chaotao Hu, Yuanqi Zhu, Qi Li, Xuanfu Ma, Jin Sun, Lingyun Chen, Tianrun Zhu, Lanyun |
| contents | Reconstructing dynamic 4D scenes from monocular videos is a fundamental yet challenging task. While recent 3D foundation models provide strong geometric priors, their performance significantly degrades in dynamic environments. This degradation stems from a fundamental tension: the inherent coupling of camera ego-motion and object motion within global attention mechanisms. In this paper, we propose a novel, training-free progressive decoupling framework that disentangles dynamics from statics in a principled, coarse-to-fine manner. Our core insight is to resolve the tension by first stabilizing the camera pose, followed by geometric refinement. Specifically, our approach consists of three synergistic components: (1) a Dynamic-Mask-Guided Pose Decoupling module that isolates pose estimation from dynamic interference, yielding a stable motion-free reference frame; (2) a Topological Subspace Surgery mechanism that orthogonally decomposes the depth manifold, safely preserving dynamic objects while injecting refined, mask-aware geometry into static regions; and (3) an Information-Theoretic Confidence-Aware Fusion strategy that formulates depth integration as a heteroscedastic Bayesian inference problem, adaptively blending multi-pass predictions via inverse-variance weighting. Extensive experiments on standard 4D reconstruction benchmarks demonstrate that our method achieves consistent and substantial improvements across principal point-cloud metrics. Notably, our approach shows competitive performance in robust 4D scene reconstruction without requiring fine-tuning, suggesting the potential of mathematically grounded dynamic-static disentanglement. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_12027 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | 4DVGGT-D: 4D Visual Geometry Transformer with Improved Dynamic Depth Estimation Zang, Ying Liu, Xuanyi Han, Yidong Ji, Deyi Ding, Chaotao Hu, Yuanqi Zhu, Qi Li, Xuanfu Ma, Jin Sun, Lingyun Chen, Tianrun Zhu, Lanyun Computer Vision and Pattern Recognition Reconstructing dynamic 4D scenes from monocular videos is a fundamental yet challenging task. While recent 3D foundation models provide strong geometric priors, their performance significantly degrades in dynamic environments. This degradation stems from a fundamental tension: the inherent coupling of camera ego-motion and object motion within global attention mechanisms. In this paper, we propose a novel, training-free progressive decoupling framework that disentangles dynamics from statics in a principled, coarse-to-fine manner. Our core insight is to resolve the tension by first stabilizing the camera pose, followed by geometric refinement. Specifically, our approach consists of three synergistic components: (1) a Dynamic-Mask-Guided Pose Decoupling module that isolates pose estimation from dynamic interference, yielding a stable motion-free reference frame; (2) a Topological Subspace Surgery mechanism that orthogonally decomposes the depth manifold, safely preserving dynamic objects while injecting refined, mask-aware geometry into static regions; and (3) an Information-Theoretic Confidence-Aware Fusion strategy that formulates depth integration as a heteroscedastic Bayesian inference problem, adaptively blending multi-pass predictions via inverse-variance weighting. Extensive experiments on standard 4D reconstruction benchmarks demonstrate that our method achieves consistent and substantial improvements across principal point-cloud metrics. Notably, our approach shows competitive performance in robust 4D scene reconstruction without requiring fine-tuning, suggesting the potential of mathematically grounded dynamic-static disentanglement. |
| title | 4DVGGT-D: 4D Visual Geometry Transformer with Improved Dynamic Depth Estimation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.12027 |