VGGT: Visual Geometry Grounded Transformer
By Jianyuan Wang, Minghao Chen, Nikita Karaev, Andrea Vedaldi, Christian Rupprecht, David Novotný. Computer Vision and Pattern Recognition, 2025
1,511 citations recorded by the source corpus.
We present VGGT, a feed-forward neural network that directly infers all key 3D attributes of a scene, including camera parameters, point maps, depth maps, and 3D point tracks, from one, a few, or hundreds of its views. This approach is a step forward in 3D computer vision, where models have typically been constrained to and specialized for single tasks. It is also simple and efficient, reconstructing images in under one second, and still outperforming alternatives that require post-processing with visual geometry optimization techniques. The network achieves state-of-the-art results in multiple 3D tasks, including camera parameter estimation, multi-view depth estimation, dense point cloud reconstruction, and 3D point tracking. We also show that using pretrained VGGT as a feature backbone significantly enhances downstream tasks, such as non-rigid point tracking and feed-forward novel view synthesis. Code and models are publicly available at https://github.com/facebookresearch/vggt.
Subfields: Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/4356b46e5dd1a4ebd579b1cd6eb3eeedddd5a65c
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
- Jiashi Feng
- Luc Van Gool
- Sergey Levine
- Trevor Darrell
- Dacheng Tao
- Jitendra Malik
- Ming-Hsuan Yang
- Abhinav Kumar Gupta
Citations in the corpus
- DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset (2024, positive): “As the quality of these models keeps improving [60, 62], we believe it would be a valuable direction to leverage these improved models for more robust camera-to-camera calibration, particularly in cluttered and low-overlap scenarios where traditional feature matching or earlier models struggle.”
- Depth Anything 3: Recovering the Visual Space from Any Views (2025, positive): “Note that using Scannet++ for training is fair to other methods, as it is widely used for training in [91, 96].”