Welcome — dataset overview and feedback

#1
by dxgl - opened

DX.GL Objaverse-1K is a collection of 1026 dense multi-view datasets rendered from curated Objaverse objects. Each dataset contains 196 views at 1024×1024 with RGB, depth, normals, masks, point clouds, and nerfstudio-compatible camera poses.

We're looking for feedback on:

  • Depth map accuracy at object edges
  • Mask quality for transparent/reflective materials
  • Camera distribution quality for your training method
  • Missing modalities or metadata

Feel free to open new discussions for bugs, requests, or results you've achieved with this data.

dxgl pinned discussion

Sign up or log in to comment