Our paper on superquadric recovery was accepted for publication in IEEE Access. In this work we propose two deep learning solutions for predicting superquadric parameters from depth images.
The paper “Learning to predict superquadric parameters from depth images with explicit and implicit supervision” is now available in early access on IEEE Xplore: https://lnkd.in/dfhwG7B
This is joint work with Tim Oblak, Jaka Šircelj, Peter Peer, Franc Solina and Aleš Jaklič.