Prof. Kot has been with
the Nanyang Technological University (NTU),
Singapore since 1991. He headed the Division
of Information Engineering at the School of
Electrical and Electronic Engineering (EEE)
for eight years. He was the Vice Dean
Research and Associate Chair (Research) for
the School of EEE for three years,
overseeing the research activities for the
School with over 200 faculty members. He was
the Associate Dean (Graduate Studies) for
the College of Engineering (COE) for eight
years. He is currently the Director of ROSE
Lab [Rapid(Rich) Object SEearch Lab) and the
Director of NTU-PKU Joint Research Institue
. He has published extensively with over 300
technical papers in the areas of signal
processing for communication, biometrics
recognition, authentication, image
forensics, machine learning and AI. He has
two USA and one Singapore patents granted.
Dr. Kot served as Associate Editor for a number of IEEE transactions, including IEEE TSP, IMM, TCSVT, TCAS-I, TCAS-II, TIP, SPM, SPL, JSTSP, JASP, TIFS, etc. He was a TC member for several IEEE Technical Committee in SPS and CASS. He has served the IEEE in various capacities such as the General Co-Chair for the 2004 IEEE International Conference on Image Processing (ICIP) and area/track chairs for several IEEE flagship conferences. He also served as the IEEE Signal Processing Society Distinguished Lecturer Program Coordinator and the Chapters Chair for IEEE Signal Processing Chapters worldwide. He received the Best Teacher of The Year Award at NTU, the Microsoft MSRA Award and as a co-author for several award papers. He was elected as the IEEE CAS Distinguished Lecturer in 2005. He was a Vice President in the Signal Processing Society and IEEE Signal Processing Society Distinguished Lecturer. He is now a Fellow of the Academy of Engineering, Singapore, a Fellow of IEEE and a Fellow of IES.
Speech Title: Skeleton-based Human Action Recognition
Abstract: In AI, action recognition is a key application. It helps in retrieving images or video based on actions like sports, fighting or hand signals. In the autonomous driving environment, it becomes very important in recognizing the hand signals from traffic police. In this research, we propose trust gates in the spatial-temporal LSTM network to improve the action recognition performance. We also propose global context-aware attention network to further improve the performance. Scale selection and feature boosting networks are also introduced to show the effectiveness in action recognition.
Jianfei is a Professor at Faculty of IT, Monash University, where he currently serves as the Head for the Data Science & AI Department. Before that, he was a full professor, a cluster deputy director of Data Science & AI Research center (DSAIR), Head of Visual and Interactive Computing Division and Head of Computer Communications Division in Nanyang Technological University (NTU). His major research interests include computer vision and multimedia. He has published 250+ technical papers in international conferences and journals. He is a co-recipient of paper awards in ACCV, ICCM, IEEE ICIP and MMSP. He has served as an Associate Editor for IEEE T-IP, T-MM, and T-CSVT as well as serving as Area Chair for ICCV, ECCV, IJCAI, ACM Multimedia, ICME and ICIP. He was the Chair of IEEE CAS VSPC-TC during 2016-2018. He had also served as the leading TPC Chair for IEEE ICME 2012 and the best paper award committee chair & co-chair for IEEE T-MM 2020 & 2019. He is a Fellow of IEEE.
Conditional Image Generation
Abstract: Since the emergence of generative adversarial networks (GAN), GAN based image generation has become more and more popular and receive a lot of attentions. In this talk, we focus on two types of GAN based image generation: image-to-image translation and image completion. For image-to-image translation, existing methods use either pixel-level cycle-consistency or feature-level matching losses, but the domain-specific nature of these losses hinder translation across large domain gaps. In contrast, we propose a novel spatially-correlative loss that is simple, efficient and yet effective for preserving scene structure consistency while supporting large appearance changes during unpaired image-to-image (I2I) translation. For image completion, unlike most image completion methods that produce only one result for each masked input, we present an approach for pluralistic image completion – the task of generating multiple and diverse plausible solutions for image completion, which is more meaningful.
Dr. Karen A. Panetta is the Dean for Graduate Education in the Tufts University School of Engineering, a Professor in the Department of Electrical & Computer Engineering and an Adjunct Professor of Computer Science and Adjunct Professor of Mechanical Engineering. She is the CEO and Co-Founder of Tessera Intelligence, which specializes in visual and sensing technology solutions for detection and recognition systems. Dr. Panetta was the 2019 President of IEEE HKN (Eta-Kappa, Nu) and is the Editor-in-Chief of the award-winning IEEE Women in Engineering Magazine. She was a NASA JOVE Fellow and Vice-President of IEEE-USA. Prior to joining the Tufts faculty, Dr. Panetta was a Principal Engineer for Digital Equipment Corporation. Karen is the recipient of the Norm Augustine Award. Previous recipients of this prestigious award included Astronaut Neil Armstrong. She is also the recipient of the Anita Borg, "Women of Vision" award and the IEEE Education Society, William Sayle Award for achievement in Engineering Education. Karen is known as a pioneer for championing engineering education to diverse and inclusive audiences. She co-authored the book "Count Girls In: Empowering Girls to Combine Any Interests with STEM to Open Up a World of Opportunity." She created the internationally acclaimed Nerd Girls Program, which promotes Science and Engineering education for young girls. has personally conducted outreach activities to over 85,000 young students, parents and community leaders around the world. She is now the co-host and creator of the web series "Nerd Girl Nation," a show that celebrates female role models using their engineering and science skills to benefit humanity. In 2011, Karen was recognized at the White House in Washington, DC by U.S. President Barack Obama and awarded the NSF Presidential (PAESMEM) Award, the nation's highest award for Excellence in Science, Mathematics and Engineering Mentoring.
Transforming Research in Image Processing
for Artificial Intelligence Applications in
Abstract: Researchers are adapting AI for a diversity of applications and they are becoming especially attractive across the healthcare fields. However, there are so many underlying concerns and challenges that need to be addressed to be able to robustly and safely trust AI for life critical applications. Much of the data relies on images and video data to develop AI driven recognition and detection systems and AI is expected to handle any quality image and perform well regardless of the image modality or quality. It is well known that if poor data is used as the input, the results of AI are not so impactful.
This talk will discuss some of the many areas AI is being used and present the the tools and methods to aid all AI researchers in their quest for efficient, reliable AI solutions by ensuring that the quality of the image data being used is well formed and can be quantitatively evalutated for use in AI systems.
Prof. Hiroshi Fujita received the B.S. and M.S. degrees in electrical engineering from Gifu University, Japan, in 1976 and 1978, respectively, and Ph.D. degree from Nagoya University in 1983. He became a research associate in 1978 and an associate professor in 1986 at Gifu National College of Technology. He was a visiting researcher at the K.Rossmann Radiologic Image Laboratory, University of Chicago, in 1983-1986. He became an associate professor in 1991 and a professor in 1995 in the Faculty of Engineering, Gifu University. He has been a professor and chair of intelligent image information since 2002 at the Graduate School of Medicine, Gifu University. He is now a Research Professor of Gifu University. He is a member of the Society for Medical Image Information (Honorary President), the Institute of Electronics, Information and Communication Engineers (Fellow), its Technical Groups on Medical Image (Adviser), the Japan Society for Medical Image Engineering (Director), and some other societies such as SPIE. He has been also served as scientific committee or program committee members, such as in International Workshop on Digital Mammography (Breast Imaging), SPIE Medical Imaging, and Computer Assisted Radiology and Surgery (CARS). He was worked as a General co-chair of Asian Forum on Medical Imaging 2007 held in Cheju National University, Korea, and as a General Chair of International Workshop for Breast Imaging (IWDM2014, Gifu). He has also worked as a Guest Editor-in-Chief in Special Section Editorial Committee for Medical Imaging, issued in April, 2013, from IEICE Society in Japan, and also as a Guest Editor-in-Chief in the Special Issue on Advanced Image Technologies in Diagnostic Imaging in 2018 in the Journal of Medical Imaging and Health Informatics. His research interests include computer-aided diagnosis system, image analysis and processing, and image evaluation in medicine. He has published over 1000 papers in Journals, Proceedings, Book chapters and Scientific Magazines.
Speech Title: Current
Status of AI Applications for Medical Image
Abstract: "Deep learning" technology in AI (artificial intelligence), which is a type of "machine learning" method in which computers learn by themselves (learning functions and rules), has reached a level where the accuracy of image recognition exceeds that of humans. Computer-aided detection/diagnosis for medical images, so-called CAD, is rapidly entering the mainstream of practical medicine. This has become part of the routine clinical task, especially for the detection/diagnosis of breast cancer with mammograms (breast imaging). In this case, the computer output is used as a "second opinion" to assist the doctor in interpreting the image. However, recent powerful AI technologies, including deep learning, highly advances the development and improvement of the performance of CAD to the next level, and it is sometimes referred to as AI-CAD. In this presentation, we will review and discuss the current state of AI-CAD, including COVID-19 diagnostic imaging, and some problems that need to be resolved for more practical use of AI-CAD in clinical settings.
References: H.Fujita, “AI-based computer-aided diagnosis (AI-CAD): The latest review to read first,” Radiological Physics and Technology, vol.13, no.1, pp.6-19, 2020.
G.Lee and H.Fujita (Eds.), “Deep Learning in Medical Image Analysis: Challenges and Applications,” Springer, 2020.