Revisiting Salient Object Detection: Simultaneous Detection, Ranking, and Subitizing of Multiple Salient Objects
This addresses the problem of inconsistent saliency judgments in computer vision, offering a more general solution for applications like image analysis, though it is incremental in refining the problem definition.
The paper tackles the ill-posed nature of salient object detection by proposing a method that simultaneously detects, ranks, and subitizes multiple salient objects, introducing new data and metrics for relative saliency. Their deep learning approach exceeds prior work across all metrics, including traditional and newly proposed ones.
Salient object detection is a problem that has been considered in detail and many solutions proposed. In this paper, 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 paper 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. 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).