Reconstructing the normal and shape at specularities in endoscopy
This addresses a domain-specific problem in medical imaging by enabling 3D perception from single endoscopic images, though it appears incremental as it builds on existing specularity analysis.
The paper tackled the problem of reconstructing tissue normal direction and curvature from specularities in endoscopic images, proposing a method that uses these features as cues for 3D perception and demonstrating results on simulated and real images.
Specularities are numerous in endoscopic images. They occur as many white small elliptic spots, which are generally ruled out as nuisance in image analysis and computer vision methods. Instead, we propose to use specularities as cues for 3D perception. Specifically, we propose a new method to reconstruct, at each specularity, the observed tissue's normal direction (i.e., its orientation) and shape (i.e., its curvature) from a single image. We show results on simulated and real interventional images.