How will Deep Learning Change Internet Video Delivery? By Dongsu Han, KAIST

Internet video has experienced tremendous growth over the last few decades and is still growing at a rapid pace. Internet video now accounts for 73% of Internet traffic and is expected to quadruple in the next five years. Augmented reality and virtual reality streaming, projected to increase twentyfold in five years, will also accelerate this trend. 

In this talk, I will argue that advances in deep neural networks present new opportunities that can fundamentally change Internet video delivery. In particular, deep neural networks allow the content delivery network to easily capture the content of the video and thus enable content-aware video delivery. To demonstrate this, I will present NAS, a new Internet video delivery framework that integrates deep neural network based quality enhancements with adaptive streaming. 

NAS incorporates a super-resolution deep neural network (DNN) and a deep re-inforcement neural network to optimize the user quality of experience (QoE). It outperforms the current state of the art, dramatically improving visual quality. It improves the average QoE by 43.08% using the same bandwidth budget or saving 17.13% of bandwidth while providing the same user QoE.  

Finally, I will talk about our recent research progress in supporting live video and mobile devices in AI-assisted video delivery that demonstrate the possibility of new designs that integrate deep learning into video delivery. 

Speaker bio: Dongsu is an associate professor at the School of Electrical Engineering at KAIST. He got his Ph.D. in Computer Science at Carnegie Mellon University. Dongsu works on transport and application layer designs in datacenter networking and cloud computing. His work has received NSDI Best Paper and Community Awards. He has served as CoNEXT 2020 TPC co-chair.