Prof. Henry Fuchs | Federico Gil Professor, University of North Carolina at Chapel Hill
Abstract: Augmented and virtual reality are hailed today as “the next big thing,” the next personal computing platform, logical successors to the previous three generations of PCs, laptops, and mobile. Others worry that today’s AR and VR systems are not yet sufficiently advanced for mass adoption, that they are more like the 1990s Apple Newton than the 2007 Apple iPhone — exciting proofs of concept, but not yet useful nor cost-effective for most consumers. This talk will review the historical development of AR and VR technologies, and survey some representative current work, sample applications, and remaining problems. Current work with encouraging results include 3D scene capture and 3D reconstruction of dynamic, populated spaces; compact and wide field-of-view AR displays; low-latency and high-dynamic range AR display systems; and near-eye lightfield displays that may reduce the vergence-accommodation conflicts that plague current AR and VR display designs.
Bio: Henry Fuchs (PhD, Utah, 1975) is the Federico Gil Distinguished Professor of Computer Science and Adjunct Professor of Biomedical Engineering at UNC Chapel Hill, coauthor of over 200 papers, mostly on rendering algorithms (BSP Trees), graphics hardware (Pixel-Planes), head-mounted / near-eye and large-format displays, virtual and augmented reality, telepresence, medical and training applications. He is a member of the National Academy of Engineering, a fellow of the American Academy of Arts and Sciences, recipient of the 2013 IEEE VGTC Virtual Reality Career Award, and the 2015 ACM SIGGRAPH Steven Anson Coons Award.
Speaker's presentation slides available HERE.
Dr. Julien Lai | Director of the Applied DL team of NVIDIA APAC Region
Abstract: One of the grand challenges of AI is to understand video content. Applications are endless: video surveillance, self-driving cars, live video streaming, ad-placement, etc. Problem is that Deep Learning, which has been boosting modern AI, is computationally expensive. It’s even more challenging when it comes to live stream video. Chip manufacturers are continuously pushing the edge of current IC technology to achieve both high throughput/performance, and high power efficiency. Engineers have also applied techniques such as network pruning, low precision, and model compression, to further accelerate network inference. That is also why our team are building the NVIDIA DeepStream SDK, which simplifies development of high performance video analytics applications powered by deep learning. We have seen successful large-scale deployment of such intelligent video analytics systems, and we see this as an unstoppable trend.
Bio: Dr. Junjie Lai received his bachelor and master degrees from Tsinghua University, and received his PhD degree from INRIA, France. His PhD research focused on GPU architecture study, performance analysis and optimization.
Dr. Lai is currently the director of the HPC software development and Applied Deep Learning team of NVIDIA APAC region. Besides leading the team, he is collaborating with developers from many well-known internet companies, to better accelerate their deep learning applications with NVIDIA GPUs. His primary focus now is to apply Deep Learning/Machine Learning techniques for various real-world problems in Computer Vision, Finance, etc.
Prof. Dah Ming Chiu | Research Professor of Chinese University of Hong Kong
Abstract: Due to recent advances in VR and AR technology and applications, more and more videos streamed on-line become 360 degree and 3D. This creates new challenges for video streaming when bandwidth is limited. Besides adapting video resolution to bandwidth, there is also opportunity to adapt what to stream to user attention, or content creator's wish for focus. This opens up exciting opportunities for interesting research. We will describe some existing work (industry seems to be leading the way in this space), and discuss our own efforts.
Bio: Dah Ming received undergraduate degree from Imperial College London, and PhD from Harvard University. After serving in industry (AT&T Bell Labs, DEC, Sun Microsystem), he joined CUHK in 2002 as a professor, and served as department chairman from 2009 to 2015.
His recent research interests include Internet content distribution, data-driven modeling and analysis of large scale systems, and network economics. He has also worked on analyzing academic social networks, and mining insights on academic research trends and evaluation methodology. Dah Ming is an IEEE Fellow. He served as an Associate Editor for IEEE/ACM Transactions on Networking from 2006 to 2011 and TPC member for many networking conferences. He was the general co-chair of ACM Sigcomm 2013 held in Hong Kong in August of 2013, with record breaking attendance.
Prof. Bo Li | Professor in the Department of Computer Science and Engineering, HKUST
Abstract: Cloud computing has taken the industry by a storm in the past decade, which, as the backend engine, has been affecting and improving our daily lives. One interesting aspect in this new computing paradigm is that communication take place not just at the edge of the system that interface with users, but plays a key role at the heart of datacenters, where large volumes of intermediate results have to be delivered to the next stage of data analytic jobs as quickly as possible, and video analytics is one of such examples with very high resource demands. In this talk, we will discuss the resource allocation for cloud computing in general, and present the case for scheduling multiple types of resources in the cloud, with the hope of supporting multiple co-existing jobs in a fair and efficient manner. With the recently introduced notion of coflow, a correlated group of flows, whose completion time is determined by the flow finishing last. We show the optimal solution relies on resolving a set of conflicting criteria, and we will illustrate some existing approaches and our endeavours along this direction.
Bio: Bo Li is a professor in the Department of Computer Science and Engineering, HKUST. He is also the Chief Technical Advisor for ChinaCache Corp. (a NASDAQ listed company), one of the leading CDN service providers in the world. He was a Cheung Kong Chair Professor in Shanghai Jiao Tong University (2010-2016).He was with IBM Networking System before joining HKUST in 1996. His recent research interests include: multimedia communications, large-scale content distribution in the Internet, datacenter networking, cloud computing, and wireless sensor networks.
He is a Fellow of IEEE. He made pioneering contributions in the Internet video streaming with a system called, Coolstreaming, which was credited as the first large-scale Peer-to-Peer live video streaming system in the world. The work received the Test-of-the-Time Paper Award from IEEE INFOCOM (2015), and USD 25M VC investment from Softbank Inc. He received the State Natural Science Award (2nd Class) and five Best Paper Awards from IEEE. He has been an editor or a guest editor for a dozen of IEEE and ACM journals and magazines. He was the Co-TPC Chair for IEEE INFOCOM 2004.
Bo Li He received his B. Eng. Degree in the Computer Science from Tsinghua University, Beijing, and his Ph.D. degree in the Electrical and Computer Engineering from University of Massachusetts at Amherst.
Speaker's presentation slides available HERE.
Measuring Subjective QoE for Interactive System Design in the Mobile Era – Lessons Learned Studying Skype Calls
Prof. Polly Huang | Professor at National Taiwan University
Abstract: As the proportion of the multimedia/game traffic over the Internet through wireless and mobile access rises and the world economy recovers slowly, the issue of streaming real-time content cost-effectively is ever more pressing. The key question to address here is how to satisfy more (paying) users given limited resources. Users switch to other services/games because they can't hear/see the content well, not because they detect the fine-grained changes in network loss, delay or jitter, so called Quality of Service (QoS). Over the years, the Internet engineers, although getting very good at designing for QoS, have overlooked the fact that users might not perceive the subtle quantitative difference in QoS metrics to the overall user experience. Towards a user-friendly, therefore economically healthy, mobile Internet, we see the need to measure, understand, and redesign various control mechanisms for quality of user experience (QoE), in addition to the QoS. Using Skype/SILK VoIP service as an example, we show how one (1) measures QoE of calls delivered of different QoS, (2) derives models that translate from QoS to QoE, and (3) exploits the model for a design that pleases the users more under the same resource constraint.
Bio: Polly is a professor at the Department of Electrical Engineering of National Taiwan University (NTU EE). Polly received her PhD from USC (CS 1999) and her BS from NTU (Math 1993). Before returning to NTU, she spent the early days of her career at AT&T Labs-Research, ETH Zurich, and UCLA. Polly was also a visiting scientist at CSAIL MIT Fall 2013. Over the past 20+ years, Polly has random walked a number of projects under the big umbrella of network and system research, including multicast routing, network simulation, Internet measurement, performance modeling, QoS, sensor networking, indoor localisation, delay-tolerant networking, time synchronisation, mobile system, event/activity inference, ubiquitous and wearable computing, e-health, and etc. The experience has nurtured her interest in design, analysis, and applications of communication networks and systems in general. Her recent interest spans the following 2 areas: multimedia networking and mobile/sensor system. Polly has (co-)authored over 100 technical articles and over 10 US patents. She has served as a TPC member for a number of high-profile network/system conferences, including ACM Sensys, Mobisys, Sigcomm, Mobicom. She has also served as an associate editor of ACM TOSN, and currently an associate editor of ACM IMWUT. She is a member of the ACM and IEEE.
Speaker's presentation slides available HERE.
Dr. Shun-Yun Hu | Co-Founder of Imonology Inc., Taiwan
Abstract: As we start to experience exponential growth in tech and productivity in certain areas, some pioneers are foreseeing a future where most society members do not need to work (in the traditional sense) for the society to support itself in most basic needs. Universal Basic Income (UBI) therefore has been proposed as a remedy to protect against the loss of job security, in a future where productivity of physical goods might far exceed demands.
Developments in Virtual Reality (VR) and Virtual Environments (VE) on the other hand, might just fill in the gap for people to find something to do and to work on, in a space that is literally limited only by humanity's imagination.
In this talk I will present both a "Good" and a "Bad" scenario of how this future might go, so that as researchers and technologists, we might start to reflect and think about how our works might impact society, knowingly or unknowingly, into humanity's bright (or dim) future. In hope that we will collectively secure ourselves a future that we all want to live in.
Shun-Yun has been doing research in scalable virtual environments using peer-to-peer (P2P) techniques since 2003, and has published his works as both research papers and free software projects (VAST and Scalra). After adventuring in Sierra games at 7 and Ultima VII later, Shun-Yun has since been fascinated by the possibilities of games and virtual worlds, and co-founded Imonology in 2010 to make virtual world creation easier and more affordable. Shun-Yun is also dad to two vegetarian kids and has a Ph.D degree in computer science from National Central University.