OmniVGGT Logo

OmniVGGT: Omni-Modality Driven Visual Geometry Grounded Transformer

1HKUST    2NTU    3SYSU    4NUS    5Alibaba Group
Equal contribution
CVPR 2026 Highlight
HKUST 南洋理工大学 sysu nus
Spatial Foundation Model Casual Prior Input 3D Reconstruction
Teaser

OmniVGGT, a model that can effectively benefit from an arbitrary number of auxiliary geometric modalities during both training and inference.

Key Contributions

OmniVGGT Framework

Overview of OmniVGGT. Our goal is to design an enhanced spatial foundation model that not only reconstructs scenes from images but also accepts an arbitrary number of corresponding camera parameters and depth information as auxiliary inputs to guide more accurate 3D scene reconstruction.

🔧 GeoAdapter for Stochastic Multimodal Inputs

OmniVGGT uses a light-weight GeoAdapter that supports arbitrary numbers of input images, auxiliary modalities, and their combinations.

📊 Stochastic Multimodal Fusion

We adopt a rule-based stochastic training strategy for OmniVGGT, ensuring its robustness and generalization when receiving varying numbers and combinations of modalities at test time.

🔌 Enhancing VLA Model Performance

We integrate OmniVGGT with vision-language-action (VLA) models. Additional auxiliary information provides the VLA model with richer spatial understanding, thereby improving its success rate across various VLA tasks.

Auxiliary Information Guidance

Compare the reconstruction quality with auxiliary information guidance.
💡 Drag the slider to compare the two models

Scene 1 / 4

Input Images

3D Reconstruction Comparison

Without Information
With Information

More Interactive Visualization

OmniVGGT reconstruction on diverse scenes. Click any thumbnail below to explore its 3D reconstruction and point cloud visualization.
💡 Hover over thumbnails to preview their image sequences

Loading 3D Model...
Meeting Room
1 / 11
💡 Controls

Experimental Results

OmniVGGT flexibly accommodates casual inputs, including arbitrary numbers of images, auxiliary information, and input combinations. It achieves state-of-the-art performance across multiple tasks, such as dense reconstruction, multi-view depth estimation, camera pose estimation, and vision-language-action, while maintaining high inference efficiency.

Main Results

Performance improvement with different ratios of auxiliary information injection.

OmniVGGT consistently outperforms VGGT even without auxiliary inputs, and all relevant metrics improve when different percentages of camera and depth information are provided.

Spatial Tracking

Visual Results of OmniVGGT with Different Auxiliary Information.

By exploiting auxiliary information, OmniVGGT is capable of accurately reconstructing complex and highly challenging geometric environments.
(Top) Camera pose and intrinsic information help correct challenging views with little or no overlap.
(Middle) Providing depth information leads to more accurate local geometry, such as on door surfaces.
(Bottom) When both depth and camera information are provided, the relative distances and viewing angles are properly corrected.

Visualization1

Visual Comparisons on 7-Scenes, NRGBD, and ETH3D datasets.
OmniVGGT exhibits accurate spatial relationships and geometric consistency, even in extremely challenging cases.

Visualization3

More Visualizations on Image Pair Input.

VLA Performance Comparison

Compare the performance of Ours and Kosmos-VLA on different tasks.
💡 Side-by-side comparison of the two methods

Page 1 / 2

Scene 1

Ours
Ours
Kosmos-VLA
Kosmos-VLA

Scene 2

Ours
Ours
Kosmos-VLA
Kosmos-VLA

Performance Table on CALVIN Benchmark

Method Tasks Completed in a Row (%) Avg. Len. ↑
1 2 3 4 5
Task: ABCD→D
Kosmos-VLA (w/ rgb) 92.9 85.4 79.4 74.4 68.1 4.00
Kosmos-VLA (w/ rgb-d) 93.4 85.8 80.5 75.3 69.2 4.04
Ours (w/ rgb) 93.8 86.6 81.0 75.5 69.5 4.07
Ours (w/ rgb-d) 93.7 86.8 81.4 76.7 70.2 4.08
Task: ABC→D
Kosmos-VLA (w/ rgb) 90.1 79.1 69.2 59.6 50.9 3.49
Kosmos-VLA (w/ rgb-d) 93.6 86.0 78.6 72.9 64.8 3.97
Ours (w/ rgb) 93.8 86.9 77.9 70.3 62.2 3.92
Ours (w/ rgb-d) 95.1 87.7 79.2 70.8 63.0 3.96

Citation

@article{peng2025omnivggt, title={OmniVGGT: Omni-Modality Driven Visual Geometry Grounded}, author={Peng, Haosong and Li, Hao and Dai, Yalun and Lan, Yushi and Luo, Yihang and Qi, Tianyu and Zhang, Zhengshen and Zhan, Yufeng and Zhang, Junfei and Xu, Wenchao and others}, journal={arXiv preprint arXiv:2511.10560}, year={2025} }