Pyramid Feature Attention Network for Saliency detection

TL;DR

This paper introduces a Pyramid Feature Attention Network that selectively emphasizes effective high-level and low-level features for improved saliency detection, incorporating attention mechanisms and an edge preservation loss.

Contribution

It proposes a novel pyramid feature extraction and attention fusion approach, along with an edge preservation loss, to enhance saliency detection accuracy.

Findings

Outperforms state-of-the-art methods on five benchmark datasets.

Effectively captures rich context and spatial features for saliency detection.

Improves boundary localization through edge preservation loss.

Abstract

Saliency detection is one of the basic challenges in computer vision. How to extract effective features is a critical point for saliency detection. Recent methods mainly adopt integrating multi-scale convolutional features indiscriminately. However, not all features are useful for saliency detection and some even cause interferences. To solve this problem, we propose Pyramid Feature Attention network to focus on effective high-level context features and low-level spatial structural features. First, we design Context-aware Pyramid Feature Extraction (CPFE) module for multi-scale high-level feature maps to capture rich context features. Second, we adopt channel-wise attention (CA) after CPFE feature maps and spatial attention (SA) after low-level feature maps, then fuse outputs of CA & SA together. Finally, we propose an edge preservation loss to guide network to learn more detailed information in boundary localization. Extensive evaluations on five benchmark datasets demonstrate that the proposed method outperforms the state-of-the-art approaches under different evaluation metrics.