A classifier-guided approach for top-down salient object detection

• A joint framework for image classification and top-down saliency estimation.
• Image classifier is used to train saliency models, and to refine the saliency map.
• Saliency maps are used to weight the image classifier for improving its accuracy.
• A computationally efficient category-aware sparse coding strategy is proposed.

We propose a framework for top-down salient object detection that incorporates a tightly coupled image classification module. The classifier is trained on novel category-aware sparse codes computed on object dictionaries used for saliency modeling. A misclassification indicates that the corresponding saliency model is inaccurate. Hence, the classifier selects images for which the saliency models need to be updated. The category-aware sparse coding produces better image classification accuracy as compared to conventional sparse coding with a reduced computational complexity. A saliency-weighted max-pooling is proposed to improve image classification, which is further used to refine the saliency maps. Experimental results on Graz-02 and PASCAL VOC-07 datasets demonstrate the effectiveness of salient object detection. Although the role of the classifier is to support salient object detection, we evaluate its performance in image classification and also illustrate the utility of thresholded saliency maps for image segmentation.