CDHN: Cross-domain hallucination network for 3D keypoints estimation

Mohammad Zohaib, Milind Gajanan Padalkar, Pietro Morerio, Matteo Taiana, Alessio Del Bue

Abstract: Highlights•We estimate keypoints from RGB images by leveraging information learnt from 3D data.•Our approach is flexible that can predict geometry based number of keypoints.•It can be trained for various categories simultaneously.•It is capable of estimating keypoints of self-occluded parts of the objects.•The estimated keypoints can be used for relative pose estimation.