Supplementary training data and binaries for 6D object pose estimation, particularly a dataset of 20 objects under various lighting conditions with RGB-D images, ground truth poses and segmentation as well as 3D models. Additionally, a collection of RGB-D images showing office backgrounds. We also provide additional pose annotations of the LINEMOD dataset (http://campar.in.tum.de/Main/StefanHinterstoisser), and Linux binaries to reproduce the results of our paper.

For more information, also see the project page: https://hci.iwr.uni-heidelberg.de/vislearn/research/scene-understanding/pose-estimation/#ECCV14