DEEP LEARNING NETWORKS OF VISUAL PROCESSING

Deep Learning Networks of Visual Processing - A growing focus of the lab is the application and development of deep learning networks as potential models of human visual performance, especially for object recognition tasks. We are currently working with a variety of networks, including AlexNet, VGG-19, GoogleNet, ResNet and Inception-v3. We have also begun constructing our own deep network architectures.

Relevant Publications

Jang, H.J., McCormack, D., Tong, F. (2017). A comparison of human performance with deep networks at recognizing objects in visual noise. Presentation at Cognitive Computational Neuroscience Conference, New York. 

Jang, H. J., & Tong, F. (2018). Can deep learning networks acquire the robustness of human recognition when faced with objects in visual noise? Talk presentation at Vision Sciences Society, St. Pete Beach, FL.