Foundation model recovers geometry and camera poses from uncalibrated images
Researchers developed a unified foundation model that recovers dense geometry, camera poses, and metric scale from uncalibrated RGB observations in single images, unordered collections, or video sequences. The model contains approximately 0.5B parameters with a shared backbone, which is less than half the size of most baselines.
1 source · Nov 17
Evidence period: Nov 16
- Hugging Face · Nov 16
Limits: AI checked each claim against excerpts from these sources. Read them for the full context.
See the sources