Place: Large Lecture Room
Abstract: In this talk, I will present our recent work on multi-level image understanding, including attribute learning, visual relationship detection and combinatorial visual semantic learning. More specifically, I will explain how we tackle the incompleteness nature of visual attributes by introducing auxiliary labels into a novel transductive learning framework, and how we propose a deep structural ranking framework for visual relationship detection to facilitate the co-occurrence of relationships and mitigate the incompleteness problem. Beyond that, I would also briefly introduce our other research topics, such as image super-resolution.
Short Bio: Dr. Hong CHANG received her PhD degree in Computer Science from Hong Kong University of Science and Technology in 2006. She was with Xerox Research Centre Europe. She is currently an Associate Researcher with the Institute of Computing Technology, Chinese Academy of Sciences. Her main research interests include algorithms and models in machine learning and their applications in computer vision, pattern recognition and data mining.