Occlusion‐robust markerless surgical instrument pose estimation
Haozheng Xu, Stamatia Giannarou · 2024
Read Occlusion‐robust markerless surgical instrument pose estimation by Haozheng Xu, Stamatia Giannarou (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11665797.txt.
Description
Haozheng Xu and Stamatia Giannarou build a vision system that tracks surgical tools even when they are partly hidden. Robot-assisted keyhole surgery needs exact tool poses for guidance and for future autonomous steps. Existing methods falter when tools are occluded or the scene shifts. Their 2024 paper proposes a markerless framework that reads the six-degree-of-freedom pose from images alone.
Tools are represented by keypoints, and a geometry solver turns those points into a stable pose. A mask-based augmentation trick trains the model to cope with hidden parts. Their test dataset carries precise ground truth from real robotic instruments. The system reached submillimeter accuracy and handled new occlusion shapes. Surgical robotics readers get robustness aimed at the messy real scene.
Catalog record
| CREATOR | Haozheng Xu, Stamatia Giannarou |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 29 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |