SU2GE-Net: a saliency-based approach for non-specific class foreground segmentation

• Published in: Scientific reports Vol. 13; no. 1; pp. 13263 - 11
• Main Authors: Lei, Xiaochun, Cai, Xiang, Lu, Linjun, Cui, Zihang, Jiang, Zetao
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 15.08.2023; Nature Publishing Group; Nature Portfolio
• Subjects: 639/166; 639/705; Computer vision; Cooperation; Deep learning; Humanities and Social Sciences; multidisciplinary; Science; Science (multidisciplinary); Segmentation; Semantics
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-023-40175-9

Salient object detection is vital for non-specific class subject segmentation in computer vision applications. However, accurately segmenting foreground subjects with complex backgrounds and intricate boundaries remains a challenge for existing methods. To address these limitations, our study proposes SU 2 GE-Net, which introduces several novel improvements. We replace the traditional CNN-based backbone with the transformer-based Swin-TransformerV2, known for its effectiveness in capturing long-range dependencies and rich contextual information. To tackle under and over-attention phenomena, we introduce Gated Channel Transformation (GCT). Furthermore, we adopted an edge-based loss (Edge Loss) for network training to capture spatial-wise structural details. Additionally, we propose Training-only Augmentation Loss (TTA Loss) to enhance spatial stability using augmented data. Our method is evaluated using six common datasets, achieving an impressive F β score of 0.883 on DUTS-TE. Compared with other models, SU 2 GE-Net demonstrates excellent performance in various segmentation scenarios.