City University of Hong Kong, Hong Kong, China
Sam Kwong received his B.Sc. degree from the State University of New York at Buffalo, M.A.Sc. in electrical engineering from the University of Waterloo in Canada, and Ph.D. from Fernuniversität Hagen, Germany. Kwong is currently a Chair Professor at the CityU Department of Computer Science, where he previously served as Department Head and Professor from 2012 to 2018. Prof. Kwong is the associate editor of leading IEEE transaction journals, including IEEE Transactions on Evolutionary Computation, IEEE Transactions on Industrial Informatics, and IEEE Transactions on Cybernetics.
He has filed more than 20 US patents, of which 13 have been granted. Kwong has a prolific research record. He has co-authored three research books, eight book chapters, and over 300 technical papers. According to Google Scholar, his works have been cited more than 25,000 times with an h-index of 70. In 2014, he was elevated to IEEE Fellow for his contributions to optimization techniques in cybernetics and video coding. He is also a fellow of Asia-Pacific Artificial Intelligence Association (AAIA) in 2022. Currently, he serves as the President of the IEEE SMC Society.
Speech Title: Intelligent Video Coding by Data-driven Techniques and Learning Models
Abstract: In June 6th 2016, Cisco released the White paper, VNI Forecast and Methodology 2015-2020, reported that 82 percent of Internet traffic will come from video applications such as video surveillance, content delivery network, so on by 2020. It also reported that Internet video surveillance traffic nearly doubled, Virtual reality traffic quadrupled, TV grew 50 percent and similar increases for other applications in 2015. The annual global traffic will first time exceed the zettabyte(ZB;1000 exabytes[EB]) threshold in 2016, and will reach 2.3 ZB by 2020. It implies that 1.886ZB belongs to video data. Thus, in order to relieve the burden on video storage, streaming and other video services, researchers from the video community have developed a series of video coding standards. Among them, the most up-to-date is the High Efficiency Video Coding(HEVC) or H.266 standard, which has successfully halved the coding bits of its predecessor, H.264/AVC, without significant increase in perceived distortion. With the rapid growth of network transmission capacity, enjoying high definition video applications anytime and anywhere with mobile display terminals will be a desirable feature in the near future. Due to the lack of hardware computing power and limited bandwidth, lower complexity and higher compression efficiency video coding scheme are still desired. For higher video compression performance, the key optimization problems, mainly decision making and resource allocation problem, shall be solved. In this talk, I will present the most recent research and new developments on deep neural network based video coding and its applications such as saliency detection, perceptual visual processing and others. This is very different from the traditional approaches in video coding. We hope applying these intelligent techniques to vide coding could allow us to go further and have more choices in trading off between cost and resources.
Prof. Xudong Jiang
Nanyang Technological University, Singapore
Xudong Jiang, IEEE Fellow, received the B.Eng. and M.Eng. from University of Electronic Science and Technology of China (UESTC), and the Ph.D. degree from Helmut Schmidt University, Hamburg, Germany. During his work in UESTC, he received two Science and Technology Awards from the Ministry for Electronic Industry of China. From 1998 to 2004, he was with the Institute for Infocomm Research, A-Star, Singapore, as a Lead Scientist and the Head of the Biometrics Laboratory, where he developed a system that achieved the most efficiency and the second most accuracy at the International Fingerprint Verification Competition in 2000. He joined Nanyang Technological University (NTU), Singapore, as a Faculty Member, in 2004, and served as the Director of the Centre for Information Security from 2005 to 2011. Currently, he is a professor of NTU. Dr Jiang holds 7 patents and has authored over 200 papers with over 40 papers in IEEE journals, including 14 papers in IEEE T-IP and 6 papers in IEEE T-PAMI. His publications are well-cited with H-index 55 and 4 of his papers were listed as the top 1% highly cited papers in the academic field of Engineering by Essential Science Indicators. He served as IFS TC Member of IEEE Signal Processing Society, Associate Editor for IEEE SPL, Associate Editor for IEEE T-IP and the founding editorial board member for IET Biometrics. Currently, Dr Jiang is an IEEE Fellow and serves as Senior Area Editor for IEEE T-IP and Editor-in-Chief for IET Biometrics. His current research interests include image processing, pattern recognition, computer vision, machine learning, and biometrics.
Speech Title: Towards Explainable AI: How Deep CNN Solves Problems of ANN
Abstract: Discovering knowledge from data has many applications in various artificial intelligence (AI) systems. Machine learning from the data is a solution to find right information from the high dimensional data. It is thus not a surprise that learning-based approaches emerge in various AI applications. The powerfulness of machine learning was already proven 30 years ago in the boom of neural networks but its successful application to the real world is just in recent years after the deep convolutional neural networks (CNN) have been developed. This is because the machine learning alone can only solve problems in the training data but the system is designed for the unknown data outside of the training set. This gap can be bridged by regularization: human knowledge guidance or interference to the machine learning. This speech will analyze these concepts and ideas from traditional neural networks to the very hot deep CNN neural networks. It will answer the questions why the traditional neural networks fail to solve real world problems even after 30 years’ intensive research and development and how the deep CNN neural networks solve the problems of the traditional neural networks and now are very successful in solving various real world AI problems.
Prof. Habib Zaidi
University of Geneva, Switzerland
Professor Habib Zaidi is Chief physicist and head of the PET Instrumentation & Neuroimaging Laboratory at Geneva University Hospital and faculty member at the medical school of Geneva University. He is also a Professor at the University of Groningen (Netherlands) and the University of Southern Denmark. His research is supported by the Swiss National Foundation, private foundations and industry (Total 8.8 M US$) and centres on hybrid imaging instrumentation (PET/CT and PET/MRI), computational modelling and radiation dosimetry and deep learning. He was guest editor for 13 special issues of peer-reviewed journals and serves on the editorial board of leading journals in medical physics and medical imaging. He has been elevated to the grade of fellow of the IEEE, AIMBE, AAPM, IOMP, AAIA and the BIR. His academic accomplishments in the area of quantitative PET imaging have been well recognized by his peers since he is a recipient of many awards and distinctions among which the prestigious (100’000$) 2010 kuwait Prize of Applied sciences (known as the Middle Eastern Nobel Prize). Prof. Zaidi has been an invited speaker of over 160 keynote lectures and talks at an International level, has authored over 360 peer-reviewed articles (h-index=69, >18’000+ citations) in prominent journals and is the editor of four textbooks.
Speech Title: New Horizons in Deep Learning-assisted Multimodality Medical Image Analysis
Abstract: Positron emission tomography (PET), x-ray computed tomography (CT) and magnetic resonance imaging (MRI) and their combinations (PET/CT and PET/MRI) provide powerful multimodality techniques for in vivo imaging. This talk presents the fundamental principles of multimodality imaging and reviews the major applications of artificial intelligence (AI), in particular deep learning approaches, in multimodality medical imaging. It will inform the audience about a series of advanced development recently carried out at the PET instrumentation & Neuroimaging Lab of Geneva University Hospital and other active research groups. To this end, the applications of deep learning in five generic fields of multimodality medical imaging, including imaging instrumentation design, image denoising (low-dose imaging), image reconstruction quantification and segmentation, radiation dosimetry and computer-aided diagnosis and outcome prediction are discussed. Deep learning algorithms have been widely utilized in various medical image analysis problems owing to the promising results achieved in image reconstruction, segmentation, regression, denoising (low-dose scanning) and radiomics analysis. This talk reflects the tremendous increase in interest in quantitative molecular imaging using deep learning techniques in the past decade to improve image quality and to obtain quantitatively accurate data from dedicated standalone (CT, MRI, SPECT, PET) and combined PET/CT and PET/MRI imaging systems. The deployment of AI-based methods when exposed to a different test dataset requires ensuring that the developed model has sufficient generalizability. This is an important part of quality control measures prior to implementation in the clinic. Novel deep learning techniques are revolutionizing clinical practice and are now offering unique capabilities to the clinical medical imaging community. Future opportunities and the challenges facing the adoption of deep learning approaches and their role in molecular imaging research are also addressed.