yangcaoai/VGGT-Det-CVPR2026

3D object detection without sensor geometry or multi-view poses

View on index · View in 3D Map

Related repos

• Video and image annotation tool for computer vision projects
• VGGT is a neural network that predicts 3D scene attributes like camera parameters, depth, and point maps from images
• Estimates camera poses and depth maps from videos
• Tracks faces and bodies in 3D using computer vision
• Tracks 3D points across multiple camera views
• Real-time object detection with DINOv3 features
• NVR with object detection for IP cameras
• Reconstructs 3D scenes from a photograph
• Model for reconstruction of 3D objects from images
• Segments objects in 3D medical images and videos