Professor Xu-Cheng Yin (殷绪成)
University of Science and Technology Beijing (北京科技大学)

Xu-Cheng Yin is a full professor and the dean of School of Computer and Communication Engineering, University of Science and Technology Beijing, China. He received the PhD degree from the Institute of Automation, Chinese Academy of Sciences, in 2006. He was a visiting professor in the College of Information and Computer Sciences, University of Massachusetts Amherst, USA, for three times (Jan 2013 to Jan 2014, Jul 2014 to Aug 2014, and Jul 2016 to Sep 2016).

殷绪成，教授、博士生导师，国家杰出青年科学基金获得者，本科、硕士毕业于北京科技大学计算机系，博士毕业于中国科学院自动化研究所，现为北京科技大学计算机与通信工程学院院长、模式识别与人工智能技术创新实验室主任，中国图象图形学学会文档图像分析与识别专委会副主任/秘书长。他长期从事模式识别、文字识别、计算机视觉及人工智能芯片技术研究及应用，在IEEE T-PAMI、IEEE T-IP、CVPR、ICDAR等学术期刊会议上发表近100篇论文；以第一完成人获北京市科技进步奖一等奖1项、教育部科技进步奖二等奖1项，带领团队连续四届（2013、2015、2017和2019年）荣获国际文档分析与识别大会技术竞赛共15项冠军

Speech Title: Open-Set Text Recognition: Definition, Framework and Algorithm
Abstract: There are a variety of novel data, novel patterns and novel labels emerged for pattern recognition and computer vision in open-set scenarios, e.g., autonomous driving and Internet multimedia understanding. In this talk, we introduce a new task, open-set text recognition, which can spot unknown (novel) characters or words, and recognize them when the corresponding side-information is provided. First, the definition and the generic framework of open-set text recognition are described. Next, a representative algorithm for open-set text recognition, i.e., character-context decoupling (separating word-contextual information and character-visual information), is reported. Finally, some potential open-set text recognition applications, e.g., text recognition for minority languages, are also discussed.

Prof. Xiaochun Cao, IET Fellow
University of Chinese Academy of Sciences, China

Xiaochun Cao is a Professor of School of Cyber Science and Technology, Sun Yat-sen University. He received the B.E. and M.E. degrees both in computer science from Beihang University (BUAA), China, and the Ph.D. degree in computer science from the University of Central Florida, USA, with his dissertation nominated for the university level Outstanding Dissertation Award. After graduation, he spent about three years at ObjectVideo Inc. as a Research Scientist. From 2008 to 2012, he was a professor at Tianjin University. Before joining SYSU, he was a professor at Institute of Information Engineering, Chinese Academy of Sciences. He has authored and coauthored over 200 journal and conference papers. In 2004 and 2010, he was the recipients of the Piero Zamperoni best student paper award at the International Conference on Pattern Recognition. He is on the editorial boards of IEEE Transactions on Image Processing and IEEE Transactions on Multimedia, and was on the editorial board of IEEE Transactions on Circuits and Systems for Video Technology.

操晓春，中山大学网络空间安全学院教授，国家杰出青年青年科学基金获得者，主要从事计算机视觉基础研究和网络空间内容安全应用研究。担任IEEE汇刊 TIP/TMM/电子学报的编委，NeurIPS/ICCV/CVPR/IJCAI/AAAI/ICPR的Area Chairs。获2019年中共中央办公厅技术进步一等奖（排名第1）。中国计算机学会优博、中国电子学会优博、中国科学院大学优博指导导师。

Speech Title: "ILL-Posed" Computer Vision Tasks

Abstract: Computer vision tasks range from the simple perspective projection matrix estimation in a traditional camera calibration application to the large-scale foundation model fitting in a contemporary object detection cloud service. One may solve most computer vision tasks through fitting functions mapping the dense, if not continuous due to quantization, visual input to a discrete and meaningful output space, including categories, bounding boxes, and depths. Due to the significant difference in cardinalities of the domain and codomain, these mapping functions fail to meet one of the three Hadamard criteria for being well-posed. In other words, the unstable computer vision solution does not depend continuously on the parameters or input data. Many researchers are trying their best to design or learn computer vision algorithms being sufficiently robust to complex perturbations such as occlusion, smoke, rain, and fog. There are also scholars looking for dedicated but powerful adversarial perturbations. Does there exist an invariant backdoor perturbation that is capable to push an arbitrary image across the decision boundary in a classification task? Are all perturbations adversarial? In this talk, I will introduce these questions our team is exploring and briefly outline some of the progress. However, much still remains unclear in spite of our efforts, and we reiterate that there might not have the answers we're looking for before AI undergoes a brand new paradigm shift.

Professor Ir Dr Yiu-Ming Cheung (FIEEE, FAAAS, FIET, FBCS)
Hong Kong Baptist University, Hong Kong

Yiu-ming Cheung (SM’06--F’18) received Ph.D. degree from Department of Computer Science and Engineering at The Chinese University of Hong Kong in 2000. He joined the Department of Computer Science in Hong Kong Baptist University (HKBU) in 2001, and became a full professor since 2012. He is an IEEE Fellow, AAAS Fellow, IET Fellow, and British Computer Society (BCS) Fellow. His research interests include Machine Learning, Intelligent Visual Computing, Pattern Recognition, Data Mining, Watermarking, and Optimization. He has published over 250 articles in the high-quality conferences and journals, including TPAMI, TNNLS, TIFS, TIP, TKDE, TCYB, CVPR, IJCAI, and AAAI. He was ranked the World’s Top 1% Most-cited Scientists in the field of Artificial Intelligence and Image Processing by Stanford University in 2020. His three co-authored papers are selected as ESI Highly Cited Papers (i.e. listed in Top 1% globally in the corresponding discipline). He has been granted one Chinese patent and two US patents. In particular, selected from 1000 new inventions and products of 700+ competition teams from 40 countries, he was awarded two most prestigious prizes: (1) the Gold Medal with Distinction (i.e. the highest grade in Gold Medals) and (2) Swiss Automobile Club Prize, in the 45th International Exhibition of Invention, Geneva, Switzerland, on March 29-April 2, 2017, in recognition of his innovative work. Also, he was the Gold Award Winner of Hong Kong Innovative Invention Award in the Seventh Hong Kong Innovative Technologies Achievement Award 2017. In addition, he won the Gold Medal with Congratulations of Jury (i.e. the highest grade in Gold Medals) at the 46th International Exhibition of Inventions of Geneva 2018. Prof. Cheung is the recipient of: (1) 2011 Best Research Award in Department of Computer Science, HKBU, (2) Best Paper Awards in SEAL’2017, ISICA’2017, ICNC-FSKD’2014, and IEEE IWDVT’2005, respectively, (3) Best Student Paper Award in ISMIS’2018, (4) Best in Theoretical Paper Award in WI-IAT’2020, and (5) 2017 IETI Annual Scientific Award. Also, he is the recipient of 2017-2019 Albert Nelson Marquis Lifetime Achievement Award. He has been elected as an IEEE CIS Distinguished Lecturer, and the Changjiang Chair Professor awarded by Ministry of Education of China. He is the Founding Chairman of IEEE (Hong Kong) Computational Intelligence Chapter, and the Chair of Technical Committee on Intelligent Informatics (TCII) of IEEE Computer Society. He has served in various capacities (e.g., Organizing Committee Chair, Program Committee Chair, Program Committee Area Chair, and Financial Chair) at several top-tier international conferences, including IJCAI’2021, ICPR’2020, ICDM’2017 & 2018, WCCI’2016, WI-IAT’2012, ICDM’2006 & WI-IAT’2006, to name a few. He is an Associate Editor for several prestigious journals, including IEEE Transactions on Neural Networks and Learning Systems, IEEE Transactions on Cybernetics, IEEE Transactions on Emerging Topics in Computational Intelligence, IEEE Transactions on Cognitive and Developmental Systems, Pattern Recognition, and Neurocomputing, as well as the Guest Editor in several international journals. Currently, he is an Engineering Panel member of Research Grants Council, Hong Kong, and a Fellow Committee member of IEEE Computational Intelligence Society.