HyKey: Hyperspectral Keypoint Detection and Matching in Minimally Invasive Surgery
Alexander Saikia, Chiara Di Vece, Zhehua Mao, Sierra Bonilla, Chloe He, Joao Ramalhinho, Tobias Czempiel, Sophia Bano, Danail Stoyanov · 2026
Read HyKey: Hyperspectral Keypoint Detection and Matching in Minimally Invasive Surgery by Alexander Saikia, Chiara Di Vece, Zhehua Mao, Sierra Bonilla, Chloe He, Joao Ramalhinho, Tobias Czempiel, Sophia Bano, Danail Stoyanov (2026) 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/arxiv-2604-17446.txt.
Description
Saikia and eight colleagues test whether hyperspectral imaging beats ordinary color cameras for finding and matching keypoints in minimally invasive surgery. Poor texture and tricky lighting defeat many RGB pipelines, so their HyKey model blends spatial and spectral cues through a hybrid 3D-2D network. Trained on a robot-collected dual-camera dataset of ex-vivo organs, it beat RGB baselines including SuperPoint and ALIKE. Mean matching accuracy reached about 97 percent, with pose accuracy about 67 percent at ten degrees, supporting steadier 3D reconstruction in surgery.
Catalog record
| CREATOR | Alexander Saikia, Chiara Di Vece, Zhehua Mao, Sierra Bonilla, Chloe He, Joao Ramalhinho, Tobias Czempiel, Sophia Bano, Danail Stoyanov |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32933 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.17446