Skip to content

M.Sc. Thesis Defense: Saliency Ranking using Deep Learning

Thursday, 24th May 2018 1:00 pm
Where: E2-461 EITC

Speaker: Mahmoud Kalash


Salient object detection is a problem that has been considered in detail and many solutions proposed. In this thesis, we argue that work to date has addressed a problem that is relatively ill-posed. Specifically, there is not universal agreement about what constitutes a salient object when multiple observers are queried.

This implies that some objects are more likely to be judged salient than others, and implies a relative rank exists on salient objects. The solution presented in this thesis solves this more general problem that considers relative rank, and we propose data and metrics suitable to measuring success in a relative object saliency landscape.

First, a novel deep learning solution is proposed based on a hierarchical representation of relative saliency and stage-wise refinement. We also show that the problem of salient object subitizing can be addressed with the same network and our approach exceeds performance of any prior work across all metrics considered (both traditional and newly proposed).

Furthermore, we present data, analysis and benchmark baseline results towards addressing the problem of salient object ranking. Methods for deriving suitable ranked salient object instances are presented, along with metrics suitable to measuring algorithm performance. In addition, we show how a derived dataset can be successively refined to provide cleaned results that correlate well with pristine ground truth. We also demonstrate the value of different rejection thresholds in determining exemplars suitable for training and evaluation, demonstrating the superior performance of a large dataset that may have some minor labeling noise over smaller extant datasets.

Finally, we provide a comparison among prevailing algorithms that address salient object ranking or detection to establish initial baselines.

Complete Seminar List
© 2011 University of Manitoba Department of Computer Science
Back to top