HyKey, a hybrid CNN trained on synthetic augmentations and epipolar constraints, achieves 96.62% mean matching accuracy and 67.18% pose estimation accuracy on ex-vivo RGB-HSI laparoscopic data, outperforming RGB baselines.
In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
HyKey: Hyperspectral Keypoint Detection and Matching in Minimally Invasive Surgery
HyKey, a hybrid CNN trained on synthetic augmentations and epipolar constraints, achieves 96.62% mean matching accuracy and 67.18% pose estimation accuracy on ex-vivo RGB-HSI laparoscopic data, outperforming RGB baselines.