Öppen gästföreläsning med Prof. Faisal Qureshi

Välkommen till en öppen gästföreläsning "A tale of two (deep) architectures: computational attention and structured generative adversarial networks" med Prof. Faisal Qureshi från University of Ontario Institute of Technology.

Banner_Faisal Qureshi2.png
Professor Faisal Qureshi

"A tale of two (deep) architectures: computational attention and structured generative adversarial network"


Abstract

I will summarize two recent projects in our lab that explore new deep learning architectures for computer vision problems.

Starting by presenting an architecture that learns to sequentially attend to different Convolutional Neural Networks (CNN) layers (i.e., “what” feature abstraction to attend to) and different spatial locations of the selected feature map (i.e., “where”) to perform the task at hand. Specifically, at each Recurrent Neural Network step, both a CNN layer and localized spatial region within it are selected for further processing. The approach is tested on two computer vision tasks: (i) image-based six degrees of freedom camera pose regression and (ii) indoor scene classification. Empirically, combining the “what” and “where” aspects of attention improves network performance on both tasks.

Second, I will present a two-stage model that separates the image inpainting problem into structure prediction and image completion. Similar to sketch art, our model first predicts the image structure of the missing region in the form of edge maps. Predicted edge maps are passed to the second stage to guide the inpainting process. This approach outperforms current state-of-the-art techniques quantitatively and qualitatively.

 

About Prof. Qureshi

Faisal Qureshi is a professor of computer science at the University of Ontario Institute of Technology. He directs the visual computing lab that focuses on problems residing at the intersection of computer vision, visual sensor networks, large-scale processing and computer graphics.

Professor Qureshi has a Ph.D. in Computer Science from University of Toronto, 2007. A M.Sc. Computer Science from University of Toronto, 2000 and a M.Sc. Electronics from Quaid-e-Azam University, Islamabad in 1995.

For more information