Keynote Speakers

Abstract

The widespread availability of high resolution aerial imagery covering wide geographical areas is spurring a revolution in large scale visual data analytics. Specifically, modern aerial wide area motion imagery (WAMI) platforms capture large high resolution at rates of 1-3 frames per second. The sequences of images, which individually span several square miles of ground area, represent rich spatio-termporal datasets that are key enablers for new applications. The effectiveness of such analytics can be enhanced by combining WAMI with alternative sources of rich geo-spatial information such as road maps or prior georegistered images. We present results from our recent research in this area covering three topics. First, we describe a novel method for pixel accurate, real-time registration of vector roadmaps to WAMI imagery based on moving vehicles in the scene. Next, we present a framework for tracking WAMI vehicles across multiple frames by using the registered roadmap and a new probabilistic framework that allows us to better estimate associations across multiple frames in a computationally tractable algorithm. Finally, in the third part, we highlight, how we can combine structure from motion and our proposed registration approach to obtain 3D georegistration for use in application such as change detection. We present results on multiple WAMI datasets, including nighttime infrared WAMI imagery, highlighting the effectiveness of the proposed methods through both visual and numerical comparisons.

Speaker Biography

Gaurav Sharma is a professor in the Departments of Electrical and Computer Engineering, Computer Science, and Biostatistics and Computational Biology, and a Distinguished Researcher in Center of Excellence in Data Science (CoE) at the Goergen Institute for Data Science at the University of Rochester. He received the PhD degree in Electrical and Computer engineering from North Carolina State University, Raleigh in 1996. From 1993 through 2003, he was with the Xerox Innovation group in Webster, NY, most recently in the position of Principal Scientist and Project Leader. His research interests include data analytics, cyber physical systems, signal and image processing, computer vision, and media security; areas in which he has 55 patents and has authored over 225 journal and conference publications. He served as the Editor-in-Chief for the IEEE Transactions on Image Processing from 2018 through 2020, and for the Journal of Electronic Imaging from 2011 through 2015. He is a member of the IEEE Publications, Products, and Services Board (PSPB), the IEEE Signal Processing Society Board of Governwors, and chaired the IEEE Conference Publications Committee in 2017-18. He is the editor of the Digital Color Imaging Handbook published by CRC press in 2003. Dr. Sharma is a fellow of the IEEE, a fellow of SPIE, a fellow of the Society for Imaging Science and Technology (IS&T) and has been elected to Sigma Xi, Phi Kappa Phi, and Pi Mu Epsilon. In recognition of his research contributions, he received an IEEE Region I technical innovation award in 2008 and the IS&T Bowman award in 2021. Dr. Sharma served as a 2020-2021 Distinguished Lecturer for the IEEE Signal Processing Society.

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. Tis 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 image analysis. 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 DL-powered 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.

Speaker Biography

Habib Zaidi is Chief physicist and head of the PET Instrumentation & Neuroimaging Laboratory at Geneva University Hospital and full Professor at the medical school of the University of Geneva. He is also a Professor at the University of Groningen (Netherlands), the University of Southern Denmark (Denmark) and Óbuda University (Hungary). His research is supported by the Swiss National Foundation, the European Commission, private foundations and industry (Total 10M+ 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 14 special issues of peer-reviewed journals and serves and serves as founding Editor-in-Chief (scientific) of the British Journal of Radiology (BJR)|Open, Deputy Editor for Medical Physics and is 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 400+ peer-reviewed articles (h-index=76, 21,500+ citations) in prominent journals and is the editor of four textbooks.

Abstract

The positioning field has huge opportunities to develop a more integrated approach to positioning technology, for example through a more organic integration of both GNSS type approaches and the rapid advances in robotics-, AI- and computer-vision driven positioning capabilities. There are also still untapped sources of inspiration for further innovation in positioning technology such as the natural world. The positioning community shares the challenges and opportunities faced by other fields like AI, in developing technology that is acceptable for widespread usage by society that adheres to our expectations and requirements around performance, introspective capability and privacy compliance. In this talk I’ll highlight some of the key areas in which we’re pursuing these goals, highlighting projects ranging from positioning technology for autonomous vehicles to the creation of new ubiquitous positioning services. I’ll touch on some of the key insights we’ve learnt on this journey and highlight exciting areas for future innovation and collaboration between disciplines and sectors.

Speaker Biography

Professor Milford conducts interdisciplinary research at the boundary between robotics, neuroscience, computer vision and machine learning, and is a multi-award winning educational entrepreneur. His research models the neural mechanisms in the brain underlying tasks like navigation and perception to develop new technologies in challenging application domains such as all-weather, anytime positioning for autonomous vehicles. From 2022 – 2027 he is leading a large research team combining bio-inspired and computer science-based approaches to provide a ubiquitous alternative to GPS that does not rely on satellites. He is also one of Australia’s most in demand experts in technologies including self-driving cars, robotics and artificial intelligence, and is a passionate science communicator. He currently holds the position of Director of the QUT Centre for Robotics, Australian Research Council Laureate Fellow, Professor at the Queensland University of Technology, and is a Microsoft Research Faculty Fellow and Fellow of the Australian Academy of Technology and Engineering.

Abstract

I started research in computer vision way back, deep in the last century (1985/86). I saw it as an exciting area: understanding how machines can process our visual world (the other major streams of AI - including how to process our linguistic world, text and speech, being somehow less inspiring.) I also was definitely attracted to the fact that it seemed an area of AI - indeed an area of Computer Science, that was “more scientific, if not more mathematical”. There were already machines that could scan documents to text, and indeed chips that could translate text to spoken word (my PhD supervisor had done work on using such technologies for teaching and aids to the “handicapped”. Yes, it was also the era before we preferred terms like “disadvantaged”.) It sort of seemed like vision was the least solved area of AI. At least to a not so well informed, about to be, PhD student. So I duly declared, with great humility, that I would essentially solve computer vision in my PhD. Five or so years later, as a slightly bruised and more humble student, I produced a thesis that arguably didn't even dent the frontier of what was known. That thesis contained the following quote in its Preface “The author is convinced that neural network like approaches are the only viable approach in the long term”. Perhaps not so prescient, after all it was the period of the second wave of enthusiasm for neural networks; and after all, it was probably more an excuse for my failure to make more of a dent. With that potted history of how my early days in Computer Vision unfolded, you can probably guess that, in the title, I mean “gone” in two different ways. One a somewhat nostalgic “where has the computer vision, as I originally perceived it, gone”, and one the more dispassionate “what were the trends, and where are we know”. In this talk I will try to suppress (but not eliminate) the nostalgia - but even so, it will necessarily be a very personal (and therefore partial, and biased) view of the field of computer vision over the approximately 40 years I have been involved in it.

Speaker Biography

David Suter graduated from The Flinders University of South Australia in 1978 with a BSc. (in applied maths and physics) and a Dip. Ed. After a couple of years of secondary school teaching, he returned to study to undertake a Grad. Dip. Computing at RMIT, and an MSc Prelim at La Trobe University. He then undertook a PhD, referred to in the above synopsis, completing in 1991. From 1988 to 1992 he was a lecturer at La Trobe University, then a Senior Lecturer (1992-2001), Associate Professor (2002-2005), and Professor (2006-2008) at Monash University. Following that, he held a Professorship at the University of Adelaide (including a stint as Head of School of Computer Science) 2008-2017. In 2018, he took up a research only Professorship at Edith Cowan University. He has served on the editorial board of several journals (Including the International Journal of Computer Vision; Pattern Recognition; Journal of Mathematical Imaging and Vision) and also on the ARC College of Experts. His main research interests include robust fitting, medical image analysis, model based segmentation, visual tracking, (and - frankly - any application for which he can obtain the necessary funding and talented PhD students and Postdocs to work on it).

Abstract

TBD.

Speaker Biography

Global Imaging AI R&D Leader and a hands-on technology executive with

• Long history of key innovations that make machines smarter leading to early contributions in signal processing, speech and document understanding, robotics, theorem proving and image understanding that resulted in over 250 refereed publications, and over 150 filed patents.
• Innovations have enabled multimillion-dollar businesses including the vacuum cleaner robots, engineering drawing scanners, syntactic theorem provers , radar frequency synthesizers, digital content management products, and Imaging AI products in health tech.
• In-depth experience spanning all aspects of industrial research, ranging from thought leadership, strategy, idea conception, research, algorithm development, R&D team development and management, commercialization into software and hardware products, and making technical sales presentations.
• Strong academic collaborations with faculty from many institutions around the world.
• Conceived the Medical Sieve Radiology Grand Challenge, a multi-million dollar global IBM research effort across IBM labs to build machines that can match the performance of an entry-level radiologist. Led to the emergence of Radiology AI as a field, and helped launch the Watson Heath Imaging business and its flagship AI products.
• Co-Chaired several international conferences over the years including CVPR 2008, IEEE HISBI 2011, IEEE ISBI 2022, and MICCAI 2023.
• Currently hold the title of IBM Fellow since 2016, the highest scientific position in IBM given to 0.01% of IBM employees. Also Fellow of IEEE, AIMBE, MICCAI, AIAA, etc.