Multiscale Deep Network with Centerness-Aware Loss for Salient Object Detection. (13th January 2022)

• Record Type: Journal Article
• Authors: Duan, Liangliang
• Other Names: Seeling Patrick Academic Editor.
• Is Part Of: Advances in multimedia. Volume 2022(2022)
• Publication Date: 2022-01-13
• Subjects: Multimedia systems -- Periodicals; Computer networks -- Periodicals; Multimédia; Réseaux d'ordinateurs; Computer networks; Multimedia systems; Periodicals; 006.7
• DOI: 10.1155/2022/2243927
• Languages: English
• ISSNs: 1687-5680

Abstract

Abstract : Deep encoder-decoder networks have been adopted for saliency detection and achieved state-of-the-art performance. However, most existing saliency models usually fail to detect very small salient objects. In this paper, we propose a multitask architecture, M2Net, and a novel centerness-aware loss for salient object detection. The proposed M2Net aims to solve saliency prediction and centerness prediction simultaneously. Specifically, the network architecture is composed of a bottom-up encoder module, top-down decoder module, and centerness prediction module. In addition, different from binary cross entropy, the proposed centerness-aware loss can guide the proposed M2Net to uniformly highlight the entire salient regions with well-defined object boundaries. Experimental results on five benchmark saliency datasets demonstrate that M2Net outperforms state-of-the-art methods on different evaluation metrics.