Image Surveillance Assistant Architecture: Status and Planned Extensions

We describe our research on integrating deep learning with artificial intelligence techniques in the context of an imagery surveillance prototype designed to automatically identify imagery of interest to a user. In particular, we briefly overview the Image Surveillance Assistant (ISA) architecture, present a study focusing on ISA1, ISA’s preliminary implementation, and discuss our plans for future extensions. Our plans include two foci: the more expressive semantic and perceptual representations and processing we will use, and user interaction processes.
Read more here: https://sites.cc.gatech.edu/~alanwags/DLAI2016/15.%20(Maynord+)%20Image%20Surveillance%20Assistant%20Architecture%20-%20Status%20and%20Planned%20Extensions.pdf