Internet video has experienced tremendous growth, with video traffic continuing to surge at a rapid pace. The continuous improvements in codecs and adaptive streaming algorithms have supported this trend.
In this talk, I will discuss how advances in deep learning unlock new opportunities that can significantly transform video delivery. To demonstrate this, I will present a series of works that integrate neural networks, thus bringing substantial quality enhancements to video in conjunction with current adaptive streaming techniques.
Bio: Dongsu Han is a professor at the School of Electrical Engineering at KAIST. He received his Ph.D. in Computer Science at Carnegie Mellon University in 2012. He has actively worked in the area of systems and networking, focusing on problems that arise from the fact that modern Internet applications often run on the cloud at scale. His current research focuses on AI-enhanced video delivery and AI systems. In the past, he was worked on user-level network stack, low-latency congestion control, and security and privacy of network applications. His impactful contributions have been featured in prestigious conferences such as SIGCOMM, OSDI, NSDI, CCS, Mobisys, and EuroSys. Notable recognitions of his work include USENIX NSDI Best Paper Award and USENIX NSDI Community Award. He is currently serving as an associate editor for IEEE/ACM Transactions on Networking and has served as a program co-chair for CoNEXT 2020.