Jun 29

1. 17:56

   Compact flat-point GPU payload → frame cap 128 (denser); GLB built from points, verified == predictions_to_glb Running

   oualidD4
   Verified signature

   cc6d487
2. 17:29

   Clarity: single trajectory line (not 64 frustums), default conf 60, honest shown/kept status

   oualidD4
   Verified signature

   1f5e38a
3. 17:19

   Sample frames across whole video (full-tour coverage); cap 64; downsample display to 2M

   oualidD4
   Verified signature

   e2130a4
4. 17:03

   Use lingbot-map-long.pt for globally-coherent reconstructions

   oualidD4
   Verified signature

   c1a519e
5. 16:45

   Viewer: elevated 3/4 default camera_position + larger viewport

   oualidD4
   Verified signature

   a7900e8
6. 16:21

   Depth->points: unproject depth (Depthmap mode); streaming model has enable_point=False

   oualidD4
   Verified signature

   06e33c8
7. 16:09

   Preprocess without torchvision (ZeroGPU base lacks it); show_error=True

   oualidD4
   Verified signature

   2de6da3
8. 15:54

   Community Gradio ZeroGPU demo of LingBot-Map streaming 3D reconstruction

   oualidD4
   Verified signature

   22c23b5
9. 15:54

   initial commit

   oualidD4
   Verified signature

   6a2d620