Deep Structured Learning
Chunhua Shen
Professor
University of Adelaide
Abstract: In this talk, I will give a brief overview of what I have been doing in terms of deep learning; and then mainly focus on two deep structured learning methods. Structured output learning concerns the problem of predicting multiple variables that have dependency, with Conditional random filed (CRF) as a typical example. It shows great promise in tasks like semantic image segmentation. Recently, there is mounting evidence that features from deep convolutional neural networks (CNN) set new records for various vision applications. Here I show how we can combine CRFs with deep CNNs to predict complex labels with considering the dependencies between the output variables. The first application is to learn depth from single monocular images. Compared with depth estimation using multiple images such as stereo depth prediction, depth from monocular images is much more challenging. We proposed a deep structured learning scheme which learns the unary and pairwise potentials of continuous CRF in a unified deep CNN framework, termed Deep Convolutional Neural Fields. For the second application, we proffer a new, efficient deep structured model learning scheme, in which we show how deep convolutional neural networks can be used to estimate the messages in message passing inference for structured prediction with CRFs. With such CNN message estimators, we obviate the need to learn or evaluate potential functions for message calculation. This confers significant efficiency for learning, since otherwise when performing structured learning for a CRF with CNN potentials it is necessary to undertake expensive inference for every stochastic gradient iteration. We also demonstrate that it yields results that are competitive with the state-of-the-art in semantic segmentation for the PASCAL VOC 2012 datasets.
Bio: Chunhua Shen is a Professor at School of Computer Science, University of Adelaide. He was awarded the ARC Future Fellowship in 2012. He is a Project Leader at the Australian Research Council Centre of Excellence for Robotic Vision, for which he leads the project on machine learning for robotic vision. He is also co-leading multiple projects in the context of Big Data at the Data to Decisions Cooperative Research Centre. These two centers receive research funding of AU$ 20M, and AU$ 25M, respectively, from the Australian commonwealth government. Before he moved to Adelaide, he was with the computer vision program at NICTA (National ICT Australia), Canberra Research Laboratory for about six years. His research interests are in the intersection of computer vision and statistical machine learning. Recent work has been on deep learning, large-scale image retrieval and classification, and scalable nonlinear optimization. He studied at Nanjing University at Australian National University, and received his PhD degree from the University of Adelaide. He is serving as Associate Editor of IEEE Transactions on Neural Networks and Learning Systems.