Mobile Edge AI: Signal/Image Processing Deep Learning Models as Real-Time Smartphone Apps

Mobile Edge AI: Signal/Image Processing Deep Learning Models as Real-Time Smartphone Apps
Speaker:  Prof. Nasser Kehtarnavaz, Univ. of Texas at Dallas

Date:   Thursday, May 26, 2022

Time:
Social Hour:  6:00 p.m.
Dinner:          6:30 p.m.
Presentation: 7:15 p.m.

Location:  The Doubletree Club Orange County Airport
7 Hutton Centre Drive,
Santa Ana, CA, 92707-5794
714-751-2400

Please RSVP at

Cost: $10 with dinner and free for presentation only.

For more information on current and future events, please visit Orange County ComSig chapter website at: http://comsig.chapters.comsoc.org

Abstract:

Edge computing solutions are estimated to grow considerably during the next few years. This talk first covers the guidelines for turning deep learning models of intelligence into apps running in real-time on smartphones as edge devices. These guidelines are then applied to a real-time image processing application and a real-time signal processing application. The image processing application involves a deep learning model to detect diabetes retinopathy in an on-the-fly manner as eye retina images are captured by smartphone cameras through commercially available lenses. The motivation behind this app is to use smartphones as edge devices to conduct cost-effective and widely accessible first-pass eye exams as an alternative to fundus cameras. The signal processing application involves implementing the speech processing pipeline of hearing aids that incorporate deep learning models. The motivation behind this app is to use smartphones as virtual hearing aid hardware platforms for the purpose of conducting hearing enhancement studies in the field or in real-world audio environments.

Speaker Biography

Prof. Nasser Kehtarnavaz is an Erik Jonsson Distinguished Professor with the Department of Electrical and Computer Engineering and the Director of the Embedded Machine Learning Laboratory at The University of Texas at Dallas, Richardson, TX. His research interests include signal and image processing, machine learning, deep learning, and real-time implementation on embedded processors. He has authored or coauthored 11 books and over 400 publications in these areas. He is a Fellow of IEEE, a Fellow of SPIE, a Fellow of AAIA, a licensed Professional Engineer, and Editor-in-Chief of Journal of Real-Time Image Processing.

Contact:  

Lan Nguyen, Ph.D.

IEEE OC ComSig Chapter Chair

lan.nguyen@linquest.com