Andrea Nicastro, Ronald Clark, Stefan Leutenegger
The paper introduces X-Section, a deep‑learning method that predicts per‑object thickness from RGB‑D images and integrates these predictions into an extended KinectFusion pipeline…