WBNet: Weakly-supervised salient object detection via scribble and pseudo-background priors

Simple item page
dc.citation.volume154
dc.contributor.authorWang Y
dc.contributor.authorWang R
dc.contributor.authorHe X
dc.contributor.authorLin C
dc.contributor.authorWang T
dc.contributor.authorJia Q
dc.contributor.authorFan X
dc.date.accessioned2024-08-08T23:49:02Z
dc.date.available2024-08-08T23:49:02Z
dc.date.issued2024-10
dc.description.abstractWeakly supervised salient object detection (WSOD) methods endeavor to boost sparse labels to get more salient cues in various ways. Among them, an effective approach is using pseudo labels from multiple unsupervised self-learning methods, but inaccurate and inconsistent pseudo labels could ultimately lead to detection performance degradation. To tackle this problem, we develop a new multi-source WSOD framework, WBNet, that can effectively utilize pseudo-background (non-salient region) labels combined with scribble labels to obtain more accurate salient features. We first design a comprehensive salient pseudo-mask generator from multiple self-learning features. Then, we pioneer the exploration of generating salient pseudo-labels via point-prompted and box-prompted Segment Anything Models (SAM). Then, WBNet leverages a pixel-level Feature Aggregation Module (FAM), a mask-level Transformer-decoder (TFD), and an auxiliary Boundary Prediction Module (EPM) with a hybrid loss function to handle complex saliency detection tasks. Comprehensively evaluated with state-of-the-art methods on five widely used datasets, the proposed method significantly improves saliency detection performance. The code and results are publicly available at https://github.com/yiwangtz/WBNet.
dc.description.confidentialfalse
dc.edition.editionOctober 2024
dc.identifier.citationWang Y, Wang R, He X, Lin C, Wang T, Jia Q, Fan X. (2024). WBNet: Weakly-supervised salient object detection via scribble and pseudo-background priors. Pattern Recognition. 154.
dc.identifier.doi10.1016/j.patcog.2024.110579
dc.identifier.eissn1873-5142
dc.identifier.elements-typejournal-article
dc.identifier.issn0031-3203
dc.identifier.number110579
dc.identifier.piiS0031320324003303
dc.identifier.urihttps://mro.massey.ac.nz/handle/10179/71246
dc.languageEnglish
dc.publisherElsevier Ltd
dc.publisher.urihttps://www.sciencedirect.com/science/article/pii/S0031320324003303
dc.relation.isPartOfPattern Recognition
dc.rights(c) 2024 The Author/s
dc.rightsCC BY 4.0
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectWeakly supervision
dc.subjectSalient object detection
dc.subjectNeural networks
dc.subjectTransformer
dc.subjectPseudo labels
dc.subjectScribble labels
dc.titleWBNet: Weakly-supervised salient object detection via scribble and pseudo-background priors
dc.typeJournal article
pubs.elements-id489036
pubs.organisational-groupOther
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Published version.pdf
Size:
4.58 MB
Format:
Adobe Portable Document Format
Description:
489036 PDF.pdf
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
9.22 KB
Format:
Plain Text
Description:
Download
Collections
Journal Articles