Learning SPD-matrix-based Representation for Visual Recognition
【讲座题目】Learning SPD-matrix-based Representation for Visual Recognition
【讲座时间】2018 年 12 月 24 日（星期一） 8:30
【主 讲 人】王雷 副教授、博士生导师、澳大利亚卧龙岗大学
Lei Wang is now Associate Professor at School of Computing and Information Technology of University of Wollongong, Australia. His research interests include machine learning, pattern recognition, and computer vision. Lei Wang has published 140+ peer-reviewed papers, including those in highly regarded journals and conferences such as IEEE TPAMI, IJCV, CVPR, ICCV and ECCV, etc. He was awarded the Early Career Researcher Award by Australian Academy of Science and Australian Research Council. He served as the General Co-Chair of DICTA 2014 and on the Technical Program Committees of 20+ international conferences and workshops. Lei Wang is senior member of IEEE.
This talk will report our recent work on learning and designing covariance and generic symmetric positive definite matrices to achieve better recognition. The first part of this talk presents a method called discriminative Stein kernel. It integrates class label information into the Stein kernel to adjust input covariance matrices to enhance its discriminative capability. The second part explores the sparsity structure among features to compute sparse inverse covariance matrix as representation, achieving better recognition performance in the case of high-dimensional features but small sample. The third part moves beyond covariance matrix to employ kernel matrix as feature representation, and jointly learn it in deep learning framework via an end-to-end manner.