Data and models for learning attention

May 3, 2016
noon - 1pm
Halligan 102
Speaker: Catherine Zhao, National University of Singapore (NUS)
Host: Roni Khardon


We develop computational and experimental methods to gain insights into visual functions and psychiatric disorders, and to build intelligent systems that approach human performance. In this talk, I will share our recent innovations on data and models, aiming at understanding and predicting visual attention in natural scenes.

I will first introduce our new approach to characterize complex scenes with rich semantics. It allows quantifying behavioral differences of multiple clinical groups. As an example, I will elaborate findings that use data and models to decipher the neurobehavioral signature of autism. I will then demonstrate an innovative psychophysical method to enable large-scale collection of attention data.

I will also present our deep learning model that effectively bridges the "semantic gap" in predicting where people look at. The model highlights semantic objects without any pre- trained detector, achieving a big leap towards human performance. Live demos will be shown to illustrate our findings and results.

Overall the integrated computational and experimental approach offers new opportunities for intelligent machines, as well as clinical and neuroscience applications. I will conclude by discussing future works in these domains.

Bio: Catherine Zhao is an assistant professor in the Department of Electrical and Computer Engineering and the Department of Ophthalmology at the National University of Singapore (NUS). She is the principal investigator at the Visual Information Processing Lab (

Catherine received the MSc and PhD degrees in computer vision from the University of California, Santa Cruz, in 2007 and 2009, respectively. Prior to joining NUS, she was a postdoctoral researcher in the Computation and Neural Systems, and Division of Biology at the California Institute of Technology from 2009 to 2011. Her main research interests include computer vision and human vision, machine learning, big data analytics, and mental disorders.

Catherine has published more than 30 journal and conference papers in top computer vision, machine learning, and cognitive neuroscience venues. She edited a book with Springer, titled Computational and Cognitive Neuroscience of Vision, which provides a systematic and comprehensive overview of vision from various perspectives, ranging from neuroscience to cognition, and from computational principles to engineering developments.