A Visual Representation-guided Framework with Global Affinity for Weakly Supervised Salient Object Detection

TL;DR

This paper introduces a visual representation-guided framework with global affinity for weakly supervised salient object detection, leveraging self-supervised learning and semantic affinity loss to improve accuracy without extra labels.

Contribution

It proposes a novel framework that combines general visual representations with task-specific features using a global semantic affinity loss, enhancing weakly supervised SOD performance.

Findings

Outperforms state-of-the-art weakly supervised SOD methods on five benchmarks.

Achieves comparable or superior results to fully supervised models.

Improves accuracy with only scribble annotations, no extra labels.

Abstract

Fully supervised salient object detection (SOD) methods have made considerable progress in performance, yet these models rely heavily on expensive pixel-wise labels. Recently, to achieve a trade-off between labeling burden and performance, scribble-based SOD methods have attracted increasing attention. Previous scribble-based models directly implement the SOD task only based on SOD training data with limited information, it is extremely difficult for them to understand the image and further achieve a superior SOD task. In this paper, we propose a simple yet effective framework guided by general visual representations with rich contextual semantic knowledge for scribble-based SOD. These general visual representations are generated by self-supervised learning based on large-scale unlabeled datasets. Our framework consists of a task-related encoder, a general visual module, and an information integration module to efficiently combine the general visual representations with task-related features to perform the SOD task based on understanding the contextual connections of images. Meanwhile, we propose a novel global semantic affinity loss to guide the model to perceive the global structure of the salient objects. Experimental results on five public benchmark datasets demonstrate that our method, which only utilizes scribble annotations without introducing any extra label, outperforms the state-of-the-art weakly supervised SOD methods. Specifically, it outperforms the previous best scribble-based method on all datasets with an average gain of 5.5% for max f-measure, 5.8% for mean f-measure, 24% for MAE, and 3.1% for E-measure. Moreover, our method achieves comparable or even superior performance to the state-of-the-art fully supervised models.