Disentangle Saliency Detection into Cascaded Detail Modeling and Body Filling

TL;DR

This paper introduces a cascaded approach to salient object detection that separates detail edge modeling from body filling, utilizing novel attention mechanisms to improve accuracy across various object sizes.

Contribution

It proposes a new two-stage framework for saliency detection with explicit detail and body modeling, along with multi-scale attention modules for better feature fusion.

Findings

Achieves state-of-the-art results on six public datasets.

Effectively captures object edges with explicit detail supervision.

Handles objects of various sizes through novel attention mechanisms.

Abstract

Salient object detection has been long studied to identify the most visually attractive objects in images/videos. Recently, a growing amount of approaches have been proposed all of which rely on the contour/edge information to improve detection performance. The edge labels are either put into the loss directly or used as extra supervision. The edge and body can also be learned separately and then fused afterward. Both methods either lead to high prediction errors near the edge or cannot be trained in an end-to-end manner. Another problem is that existing methods may fail to detect objects of various sizes due to the lack of efficient and effective feature fusion mechanisms. In this work, we propose to decompose the saliency detection task into two cascaded sub-tasks, \emph{i.e.}, detail modeling and body filling. Specifically, the detail modeling focuses on capturing the object edges by supervision of explicitly decomposed detail label that consists of the pixels that are nested on the edge and near the edge. Then the body filling learns the body part which will be filled into the detail map to generate more accurate saliency map. To effectively fuse the features and handle objects at different scales, we have also proposed two novel multi-scale detail attention and body attention blocks for precise detail and body modeling. Experimental results show that our method achieves state-of-the-art performances on six public datasets.