{"type":"ScholarlyArticle","title":"G-SalAlignMamba: Geometry-Aware Vision Mamba for Dual-Modal Salient Object Detection","authors":[{"name":"Haixiao Gao","slug":"haixiao-gao","given":"Haixiao","family":"Gao","affiliation":""},{"name":"Yimin Zheng","slug":"yimin-zheng","given":"Yimin","family":"Zheng","affiliation":""},{"name":"Mengke Song","slug":"mengke-song","given":"Mengke","family":"Song","affiliation":""},{"name":"Linyou Xiao","slug":"linyou-xiao","given":"Linyou","family":"Xiao","affiliation":""},{"name":"Tian-Tian Zhang","slug":"tian-tian-zhang","given":"Tian-Tian","family":"Zhang","affiliation":""},{"name":"Zhi-Ri Tang","slug":"zhi-ri-tang","given":"Zhi-Ri","family":"Tang","affiliation":""}],"abstract":"Recently, Visual State Space Models offer powerful global modeling for Dual-modal Salient Object Detection (SOD). However, they are still constrained by three inherent limitations: first, Mamba's strict reliance on sequential ordering makes it sensitive to cross-modal geometric misalignment, where spatial shifts disrupt token correspondence; second, general indiscriminate scanning treating all tokens equally may lead to signal dilution, where sparse foreground features are overwhelmed by background noise; third, conventional decoders rely on implicit upsampling, causing boundary degradation during resolution recovery. To address these challenges, we propose G-SalAlignMamba, a geometry-aware framework tailored for dual-modal SOD. We introduce Geometry-Aware Encoding with explicit alignment to correct spatial shifts, Semantics-Informed Refinement to prevent signal dilution by prioritizing foregrounds, and Structure-Preserving Decoding that integrates explicit alignment with unsupervised boundary refinement. Extensive experiments show that G-SalAlignMamba achieves state-of-the-art performance on RGB-D and RGB-T benchmarks with favorable efficiency (30.41 FPS, 83.80M parameters). The code is available at https://github.com/PC1-99/G-SalAlignMamb.git.","doi":"10.24963/ijcai.2026/123","url":"https://ijcai.org/proceedings/2026/123","pdf":"https://ijcai.org/proceedings/2026/0123.pdf","markdown":"https://ijcai.org/proceedings/2026/123.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/123","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"Computer Vision","track":"Main Track","number":123,"pages":{"first":1097,"last":1105},"keywords":["Computer Vision: Multimodal learning","Computer Vision: Scene analysis and understanding"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}