Dr. Amy Reibman
AT&T Labs – Research
Florham Park, NY

Digital image processing is everywhere today: digital photography, digital radiology, digital cinema, video conferencing, and streaming video on the web. An accurate method to compare the quality of images and video is necessary so that algorithms can be optimized, products can be benchmarked, video outages can be detected, and service-level agreements can be written. Unfortunately, the complexity of the human visual system makes accurate assessment challenging.

Peak Signal-to-Noise ratio (PSNR) (or equivalently mean-squared error (MSE)) is a simple measure of image quality based on average error. Even in the early days of digital image processing in the 1960's, it was known that PSNR was a poor predictor of image quality. This has spurred decades of research into improved methods of image quality assessment. Despite its drawbacks, however, PSNR is still heavily reported. Furthermore, researchers have also tried to incorporate PSNR into their optimization algorithms -- can this actually provide gains?

This talk provides a broad overview of objective methods for image and video quality assessment. We give visual examples and describe scenarios in which PSNR is misleading, inappropriate, or completely inapplicable. We also describe scenarios in which PSNR has proved very effective, where dramatic visual improvements in image quality can be achieved with its use. Finally, we present a sampling of alternate approaches to characterize image and video quality, including our recent contributions on measuring video quality inside the network.

Amy R. Reibman is a Distinguished Lecturer in the IEEE Signal Processing Society. She received the B.S., M.S. and Ph.D. degrees in electrical engineering from Duke University in 1983, 1984, and 1987, respectively. From 1988 to 1991, she was an assistant professor in the Department of Electrical Engineering at Princeton University. In 1991 she joined AT&T Bell Laboratories, and became a Distinguished Member of Technical Staff in 1995. She is currently a Lead Member of Technical Staff in the Communication Sciences and Artificial Intelligence Research Department at AT&T Laboratories.

Dr. Reibman was elected IEEE Fellow in 2005, for her contributions to video transport over networks. In 1998, she won the IEEE Communications Society Leonard G. Abraham Prize Paper Award. She was the Technical co-chair of the IEEE International Conference on Image Processing in 2002; the Technical Co-chair for the First IEEE Workshop on Multimedia Signal Processing in 1997; the Technical Chair for the Sixth International Workshop on Packet Video in 1994.

Dr. Reibman's research interests include video quality estimation for video compression, transport over packet and wireless networks, superresolution image and video enhancement, and 3-D and multiview video.