🧱 brickmix scanner

Ready

VGGT Pipeline

VGGT (CVPR 2025 Best Paper) — Feed-forward 3D reconstruction from multi-view images. Produces point clouds, depth maps, and camera poses in seconds.

📷 Capture → 🔧 Preprocess → 🧠 VGGT → ☁️ Point Cloud → 🖥️ Render

Images

Points

Inference (s)

Input Images

Click Capture to scan with turntable, or images will load from last scan.

3D Point Cloud Viewer

Run reconstruction to see 3D point cloud

Log

[VGGT] Visual Geometry Grounded Transformer — CVPR 2025 Best Paper

[VGGT] Model: facebook/VGGT-1B (1B parameters)

[VGGT] Ready. Use Capture → Reconstruct pipeline.