# Region Refinement Network for Salient Object Detection

## TL;DR

This paper introduces a Region Refinement Network with a novel refinement module and boundary loss that improve salient object detection accuracy by reducing false predictions and sharpening boundaries, while maintaining efficiency.

## Contribution

The paper proposes a new Region Refinement Module and Boundary Refinement Loss that enhance saliency detection accuracy and boundary clarity with minimal model size increase.

## Key findings

- Significant reduction in false predictions from background.
- Improved boundary accuracy in saliency maps.
- Model generalizes well to related tasks like portrait segmentation.

## Abstract

Albeit intensively studied, false prediction and unclear boundaries are still major issues of salient object detection. In this paper, we propose a Region Refinement Network (RRN), which recurrently filters redundant information and explicitly models boundary information for saliency detection. Different from existing refinement methods, we propose a Region Refinement Module (RRM) that optimizes salient region prediction by incorporating supervised attention masks in the intermediate refinement stages. The module only brings a minor increase in model size and yet significantly reduces false predictions from the background. To further refine boundary areas, we propose a Boundary Refinement Loss (BRL) that adds extra supervision for better distinguishing foreground from background. BRL is parameter free and easy to train. We further observe that BRL helps retain the integrity in prediction by refining the boundary. Extensive experiments on saliency detection datasets show that our refinement module and loss bring significant improvement to the baseline and can be easily applied to different frameworks. We also demonstrate that our proposed model generalizes well to portrait segmentation and shadow detection tasks.