Scalable Permutation-Equivariant Visual Geometry Learning
VGGT (CVPR 2025)
The demo for pixel reasoner
Generate 3D point cloud from input image