Title: Harnessing Edge Intelligence for Video Services

Speaker Name: Dr. Suvadip Batabyal

Speaker Email: suvadip.batabyal@cse.nitdgp.ac.in

Speaker Affiliation:  Assistant Professor, Department of CSE, NIT Durgapur


Abstract: Edge devices can not only reduce the latency of big-data applications, but can also help in making intelligent decisions in a real-time manner. This is made possible due to the proliferation of a large number of heterogeneous edge devices being in proximity of the data source. One of the several applications of employing edge intelligence is provisioning of video services. Video services include on-demand and real-time video streaming and video surveillance. The viewing experience, measured using quality of experience (QoE), for streaming services and the video analytics for surveillance systems are largely affected by the quality of the received video, which in turn depends on the network parameters. Edge intelligence can aid in delivering video services by (i) caching video data (ii) dynamic bit-rate adaptation (iii) offloading computation. The presentation will discuss how edge intelligence, models and supports different video services. We will first discuss the architecture of multi-edge access computing (MEC) and service models. Then we will briefly look into different video coding techniques, which will help in understanding the concept of quality of experience (QoE). Finally, we will see how intelligence can be built into MEC and how it aids in enhancing the delivery of video services. The presentation will be concluded by highlighting the state of ongoing research and open challenges in this area.

LINK FOR REGISTRATION: https://forms.gle/KFm8RW1SBawK9QiHA

Date: 14th June 2023

Time: 7:00-8:00 PM (IST)

Mode: Online