Local Context Attention for Salient Object Segmentation

TL;DR

This paper introduces LCANet, a novel network for salient object segmentation that leverages local context attention through specialized modules, achieving superior accuracy on benchmark datasets.

Contribution

The paper proposes a new Local Context Attention Network with an Attentional Correlation Filter and a coarse-to-fine structure for improved salient object segmentation.

Findings

Achieves 0.883 max F-score on DUTS-TE dataset.

Outperforms state-of-the-art methods in salient object segmentation.

Demonstrates effective local context modeling through novel modules.

Abstract

Salient object segmentation aims at distinguishing various salient objects from backgrounds. Despite the lack of semantic consistency, salient objects often have obvious texture and location characteristics in local area. Based on this priori, we propose a novel Local Context Attention Network (LCANet) to generate locally reinforcement feature maps in a uniform representational architecture. The proposed network introduces an Attentional Correlation Filter (ACF) module to generate explicit local attention by calculating the correlation feature map between coarse prediction and global context. Then it is expanded to a Local Context Block(LCB). Furthermore, an one-stage coarse-to-fine structure is implemented based on LCB to adaptively enhance the local context description ability. Comprehensive experiments are conducted on several salient object segmentation datasets, demonstrating the superior performance of the proposed LCANet against the state-of-the-art methods, especially with 0.883 max F-score and 0.034 MAE on DUTS-TE dataset.