Robust Salient Object Detection on Compressed Images Using Convolutional Neural Networks

TL;DR

This paper benchmarks and analyzes CNN-based salient object detection on compressed images, revealing robustness issues and proposing a new framework to improve performance under image compression artifacts.

Contribution

It provides a comprehensive benchmark, analysis, and a simple baseline framework to enhance CNN-based SOD robustness on compressed images.

Findings

State-of-the-art CNN SOD models perform poorly on compressed images.

Robustness is mainly affected by compressed image characteristics and feature learning limitations.

The proposed framework improves robustness while maintaining accuracy on clean images.

Abstract

Salient object detection (SOD) has achieved substantial progress in recent years. In practical scenarios, compressed images (CI) serve as the primary medium for data transmission and storage. However, scant attention has been directed towards SOD for compressed images using convolutional neural networks (CNNs). In this paper, we are dedicated to strictly benchmarking and analyzing CNN-based salient object detection on compressed images. To comprehensively study this issue, we meticulously establish various CI SOD datasets from existing public SOD datasets. Subsequently, we investigate representative CNN-based SOD methods, assessing their robustness on compressed images (approximately 2.64 million images). Importantly, our evaluation results reveal two key findings: 1) current state-of-the-art CNN-based SOD models, while excelling on clean images, exhibit significant performance bottlenecks when applied to compressed images. 2) The principal factors influencing the robustness of CI SOD are rooted in the characteristics of compressed images and the limitations in saliency feature learning. Based on these observations, we propose a simple yet promising baseline framework that focuses on robust feature representation learning to achieve robust CNN-based CI SOD. Extensive experiments demonstrate the effectiveness of our approach, showcasing markedly improved robustness across various levels of image degradation, while maintaining competitive accuracy on clean data. We hope that our benchmarking efforts, analytical insights, and proposed techniques will contribute to a more comprehensive understanding of the robustness of CNN-based SOD algorithms, inspiring future research in the community.