Contemporary Visual Computing: A System Perspective Chang-Wen Chen Abstract: Visual computing,
traditionally, is a generic term for all computer science disciplines dealing
with images, videos, and other types of visual data. These disciplines mainly
include computer graphics, image processing, visualization, computer vision,
virtual and augmented reality, and video analytics. This talk shall analyze
contemporary visual computing systems from several systematic perspectives.
First, contemporary visual data acquisition has shifted from the laboratory in
the early days to the fields in recent years with new technical challenges
emerging on the visual sensing front. Second, massive visual data acquired for
a very diverse range of applications require high-performance computation of
visual data via cloud computing. Such extension of visual computing to both
ends of the contemporary system now demands pervasive networking to effectively
transport such volumetric visual data back and forth. Therefore, the networking
of visual data has now become a key component in contemporary visual computing
systems which has not been adequately studied before. The investigation of
visual computing systems now needs to be significantly deepened to facilitate
the researchers to traverse across new domains of exploitation. Several
examples of emerging applications with unique design principles will be
presented to illustrate the technical challenges we are facing and the
potential impacts that contemporary visual computing systems are capable of
producing. Bio: Prof. Changwen Chen is currently Chair Professor of Visual Computing at
The Hong Kong Polytechnic University. Before his current position, he served as
Dean of the School of Science and Engineering at The Chinese University of Hong
Kong, Shenzhen from 2017 to 2020, and concurrently as Deputy Director at Peng
Cheng Laboratory from 2018 to 2021. Previously, he has been an Empire
Innovation Professor at the State University of New York at Buffalo from 2008
to 2021 and the Allan Henry Endowed Chair Professor at the Florida Institute of
Technology from 2003 to 2007. He has served as an Editor-in-Chief for T-MM
(2014-2016) and T-CSVT (2006-2009) and has received many professional
achievement awards, including the prestigious Alexander von Humboldt Award in
2010, SUNY Chancellor’s Award for Excellence in Scholarship and Creative
Activities in 2016, and UIUC ECE Distinguished Alumni Award in 2019. He is an
IEEE Fellow, an SPIE Fellow, and a Member of the Academia Europaea. The future of
video communication
Baining Guo Abstract: I will talk about
what the future video communication system might look like, covering both the
front-end, the back-end and some fundamental theoretical issues. For the front
end, I will talk about 3D immersive video communication. For the back end, I will
talk about neural codec, which is on its way to replace our traditional codec.
For theoretical issues, I will touch on both the fundamental changes brought by
the Transformer in the image and video processing domain and some fundamental
questions that we are only begin to ponder. Bio: Dr. Baining
Guo is a Distinguished Scientist of Microsoft Corporation. His main research
areas are computer graphics, geometric modeling, virtual reality, and computer
vision. Prior to joining Microsoft Research in 1999, he was a senior staff
researcher with Intel Research in the Silicon Valley. Dr. Guo got his BS degree
from Beijing University and his MS and PhD degrees from Cornell University. Dr.
Guo is an IEEE fellow and ACM fellow. He is also a member of Canadian Academy
of Engineering. New frontiers in
machine learning interpretability
Mihaela van der Schaar Abstract: Medicine has the potential to be transformed by
machine learning (ML) by addressing core challenges such as time-series
forecasts, clustering (phenotyping), and heterogeneous treatment effect
estimation. However, to be embraced by clinicians and patients, ML approaches
need to be interpretable. So far though, ML interpretability has been largely
confined to explaining the predictions of static classifiers. In
this keynote, I describe an extensive new framework for ML interpretability.
This framework allows us to 1) interpret ML methods for time-series
forecasting, clustering (phenotyping), and heterogeneous treatment effect
estimation using feature and example-based
explanations, 2) provide personalized explanations of ML methods with reference
to a set of examples freely selected by the user, and 3) autonomously
(re)discover known scientific concepts using concept activation regions, which
are generalizations of concept-based explanations. To learn more about our work in this area - see our website
dedicated to this topic - https://www.vanderschaar-lab.com/interpretable-machine-learning/ and our github
- https://github.com/vanderschaarlab/Interpretability Bio: https://www.vanderschaar-lab.com/prof-mihaela-van-der-schaar/ The trend of the
post deep learning era: foundation models
Dacheng Tao JD Explore Academy Bio: Dacheng Tao is
the Inaugural Director of the JD Explore Academy and a Senior Vice President of
JD.com. He is also an advisor and chief scientist of the digital science
institute in the University of Sydney. He mainly applies statistics and
mathematics to artificial intelligence and data science, and his research is
detailed in one monograph and over 200 publications in prestigious journals and
proceedings at leading conferences. He received the 2015 Australian
Scopus-Eureka Prize, the 2018 IEEE ICDM Research Contributions Award, and the
2021 IEEE Computer Society McCluskey Technical Achievement Award. He is a
fellow of the Australian Academy of Science, the World Academy of Sciences, the
Royal Society of NSW, AAAS, ACM, IAPR and IEEE.
The
Hong Kong Polytechnic University
Microsoft
Research
University
of Cambridge