Lune

CVPR2026Top-tier venue

Uncertainty-Aware Modality Fusion for Unaligned RGB-T Salient Object Detection

Mianzhao Wang, Fan Shi, Xu Cheng, Chen Jia, Shengyong Chen

Abstract

Unaligned RGB-T salient object detection (SOD) remains challenging due to severe cross-modal spatial discrepancies and unreliable feature fusion. Existing methods often assume perfect alignment or rely on geometric registration, which is computationally demanding and sensitive to cross-modal inconsistencies. To address these limitations, we propose an uncertainty-aware modality fusion network (UMFNet) that reformulates RGB-T SOD as an uncertainty-aware representation learning problem. Specifically, the proposed uncertainty alignment module (UAM) models pixel-wise features as Gaussian latent distributions to estimate local uncertainty and identify cross-modal consistency regions within the feature space, thereby achieving implicit alignment without explicit registration. Furthermore, the confidence-guided global modulation (CGM) mechanism leverages confidence maps derived from uncertainty estimation to adaptively regulate the fusion of RGB and thermal features, enhancing salient cues in reliable regions while suppressing noisy or inconsistent information. Extensive experiments on five unaligned and three aligned RGB-T SOD benchmarks demonstrate that UMFNet achieves state-of-the-art performance across diverse alignment conditions.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • VSCode: General Visual Salient and Camouflaged Object Detection with 2D Prompt Learning cites Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
  • Alignment-Free RGB-T Salient Object Detection: A Large-Scale Dataset and Progressive Correlation Network cites Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on6

  1. Swin Transformer: Hierarchical Vision Transformer using Shifted WindowsZe Liu, Yutong Lin, Yue Cao, Han Hu et al.ICCV 2021 · 31,683 citations
  2. Selectivity or Invariance: Boundary-Aware Salient Object DetectionJinming Su, Jia Li, Yu Zhang, Changqun Xia et al.ICCV 2019 · 192 citations
  3. VSCode: General Visual Salient and Camouflaged Object Detection with 2D Prompt LearningZiyang Luo, Nian Liu, Wangbo Zhao, Xuguang Yang et al.CVPR 2024 · 96 citations
  4. Backdoor Attacks on Bimodal Salient Object Detection with RGB-Thermal DataWen Yin, Bin Benjamin Zhu, Yulai Xie, Pan Zhou et al.ACM MM 2024 · 1 citation
  5. Alignment-Free RGB-T Salient Object Detection: A Large-Scale Dataset and Progressive Correlation NetworkKunpeng Wang, Keke Chen, Chenglong Li, Zhengzheng Tu et al.AAAI 2025 · 1 citation

Related papers

  1. IGIANet: Illumination Guided Implicit Alignment Network for Infrared-Visible UAV DetectionXiangqi Chen, Dawei Zhang, Li Zhao, Chengzhuan Yang et al.AAAI 2026
  2. DiMSOD: A Diffusion-Based Framework for Multi-Modal Salient Object DetectionShuo Zhang, Jiaming Huang, Wenbing Tang, Yan Wu et al.AAAI 2025 · 3 citations
  3. Breaking Alignment Barriers: TPS-Driven Semantic Correlation Learning for Alignment-Free RGB-T Salient Object DetectionLupiao Hu, Fasheng Wang, Fangmei Chen, Fuming Sun et al.AAAI 2026
  4. Attentive Alignment Network for Multispectral Pedestrian DetectionNuo Chen, Jin Xie, Jing Nie, Jiale Cao et al.ACM MM 2023 · 26 citations
  5. MMNet: Multi-Stage and Multi-Scale Fusion Network for RGB-D Salient Object DetectionGuibiao Liao, Wei Gao, Qiuping Jiang, Ronggang Wang et al.ACM MM 2020 · 53 citations
Dusk over the sea between two cliffs drawn in fine vertical lines