We investigate the possibility of providing a confidence score to computational saliency maps. This research focus on studying saliency maps as data structures and we try to identifying common features in such structures that can help in distinguishing reliable saliency information. The end goal of this project is to propose a robust algorithm that can provide confidence level of each pixel in a saliency map given the input video frame and generated saliency map.
- T. Alshawi, Z. Long, and G. AlRegibb, “Unsupervised Estimation Of Uncertainty For Video Saliency Detection Using Temporal Cues” IEEE Global Conf. on Signal and Information Processing (GlobalSIP), Orlando, Florida, Dec. 14-16, 2015. [PDF] [PPT (Poster/Slide)] [Bib] [Code]
T. Alshawi, Z. Long, and G. AlRegib, “Unsupervised Uncertainty Analysis For Video Saliency Detection” the 49th Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, Nov. 8-11, 2015. [PDF] [PPT (Poster/Slide)] [Bib] [Code]