Co-saliency detection via partially absorbing random walk
Xing Sun, Lihe Zhang, Huchuan Lu · 2017
Co-saliency detection aims at finding the common salient objects in multiple images. In this paper, we introduce a new co-saliency detection model, which includes two main parts: co-salient seed selection using the inter-object recurrence cues from multiple images and saliency label propagation using partially absorbing random walk. With the guidance of co-salient seeds, salient objects are individually detected from each image through a semi-supervised label propagation process. Experimental results on two benchmark databases demonstrate that the proposed method achieves good performance.