This chapter reviews the rich literature of descriptors, which have been proposed for the analysis of the shape of 3D models, and the dissimilarity measures, which have been used to compare them. It focuses on global descriptors, which are grouped into four different categories, namely, distribution-based, view-based, spherical representation-based, and deep neural network-based. In the recent literature, deep neural networks have revolutionized image analysis. They have been used to learn general-purpose image descriptors which outperform hand-crafted features on a number of vision tasks such as object detection, texture and scene recognition, and fine-grained classification. The most straightforward way to extend these techniques to the 3D case is by using a multiview representation of the 3D objects. A Convolutional Neural Network (CNN) takes an image as an input and produces a set of scores where each score is the likelihood that the image belongs to one of the image classes.