Leveraging Human Fixations in Sparse Coding: Learning a Discriminative Dictionary for Saliency Prediction.

Ming Jiang,Mingli Song,Qi Zhao
DOI: https://doi.org/10.1109/smc.2013.364
2013-01-01
Abstract:This paper proposes to learn a discriminative dictionary for saliency detection. In addition to the conventional sparse coding mechanism that learns a representational dictionary of natural images for saliency prediction, this work uses supervised information from eye tracking experiments in training to enhance the discriminative power of the learned dictionary. Furthermore, we explicitly model saliency at multi-scale by formulating it as a multi-class problem, and a label consistency term is incorporated into the framework to encourage class (salient vs. non-salient) and scale consistency in the learned sparse codes. K-SVD is employed as the central computational module to efficiently obtain the optimal solution. Experiments demonstrate the superior performance of the proposed algorithm compared with the state-of-the-art in saliency prediction.
What problem does this paper attempt to address?