Foreword to the Special Section on Eurographics Workshop on 3D Object Retrieval 2017

The special section of Computers & Graphics (C&G), featuring the selected best papers presented at the 10th Eurographics Workshop on 3D Object Retrieval - 3DOR 2017 which was held on April 23-24, 2017 in Lyon, France. In the first paper, '3D point cloud semantic labeling with 2D deep segmentation networks.' Alexandre Boulch, Joris Guerry, Bertrand Le Saux, and Nicolas Audebert investigate the use of deep learning for semantic labeling of unstructured point clouds. Another paper titled 'Towards real- time 3D object recognition: A lightweight volumetric CNN framework using multitask learning' is by Shuaifeng Zhi, Yongxiang Liu, Xiang Li, and Yulan Guo . The authors use volumetric convolutional neural networks for real-time 3D object recognition. The proposed architecture combines two learning tasks and allows to obtain state-of-the-art 3D object recognition performance in real-time with the smallest number of training parameters. The last paper, 'Ensemble of PANORAMA-based Convolutional Neural Networks for 3D Model Classification and Retrieval', from Konstantinos Sfikas, Ioannis Pratikakis, and Theoharis Theoharis, also presents a deep learning approach for 3D object recog- nition. 2D Convolutional Neural Networks are applied on 2D panoramic view representations of 3D models.