Many tasks in Pattern Recognition and Document Image Analysis are formulated as graph matching problems. Despite the NP-hard nature of the problem, fast and accurate approximations have led to significant progress in a wide range of applications in pattern recognition. Therefore learning graph-based representations and related techniques is a real interest of the community. In this tutorial, we will present many methodologies for obtaining stable graph representation for different applications. Afterwards, we will explain different graph-based algorithms, methods and techniques for performing recognition, classification, detection, and many other tasks in graph domain. We will present the recent trends including the graph convolutional networks and message passing in graphs highlighting the applications to various pattern recognition problem such as chemical molecule classification and community detection in graph representation of social networks. Moreover, in addition to different applications of these algorithms in the field of Document Image Analysis & Recognition in particular and Pattern Recognition in general, a hands-on experience for working with graphs will also be provided.
Structural Pattern Recognition, Graph-based representations, Graph matching, Graph embedding, Graph kernel, Graph serialization, Graph indexing, Graph hashing, Subgraph spotting and Graph Neural Network.