Research map / Vision-language and multimodal / Cross-modal learning
SMLNet: A SPD Manifold Learning Network for Infrared and Visible Image Fusion
Huan Kang, Hui Li, Tianyang Xu, Xiaojun Wu, Rui Wang, Chunyang Cheng, Josef Kittler
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use cross-modal covariance fusion instead of infrared-only single-modal (ablation) | SMLNet · TNO | Q^AB/F: 0.426 -> 0.516 | Improved |
| use cross-modal covariance fusion instead of visible-only single-modal (ablation) | SMLNet · TNO | Q^AB/F: 0.449 -> 0.516 | Improved |
| use 1 BiMap/ReEig block instead of 2 (ablation) | SMLNet · TNO | Q^AB/F: 0.449 -> 0.516 | Improved |
| use 1 BiMap/ReEig block instead of 3 (ablation) | SMLNet · TNO | SF: 9.784 -> 11.098 | Improved |
| use SPDAM manifold attention instead of Dense Block (ablation) | SMLNet · TNO | Q^AB/F: 0.395 -> 0.516 | Improved |
| use SPDAM instead of CBAM (ablation) | SMLNet · TNO | Q^AB/F: 0.378 -> 0.516 | Improved |
| use SPDAM instead of ViT (ablation) | SMLNet · TNO | SF: 9.069 -> 11.098 | Improved |
| keep orthogonality constraint on BiMap weights (ablation) | SMLNet · TNO | VIF: 0.772 -> 0.832 | Improved |
| regularization epsilon 0.001 instead of 0.0001 (ablation) | SMLNet · TNO | VIF: 0.504 -> 0.832 | Improved |
| regularization epsilon 0.001 instead of 0.01 (ablation) | SMLNet · TNO | SF: 10.164 -> 11.098 | Improved |
| SPD manifold learning instead of PCA reduction (ablation) | SMLNet · MSRS manifold distribution | CM-NNR / SS: CM-NNR 0.375 -> 0.442; SS 0.045 -> 0.016 | Improved |
| SPDAM instead of Dense Block under Gaussian noise (ablation) | SMLNet · TNO (Gaussian noise) | SD: 36.238 -> 42.199 | Improved |
Closest papers by meaning
- Mining on Heterogeneous Manifolds for Zero-Shot Cross-Modal Image Retrieval (AAAI 2020 · similarity 0.86)
- Dimensionality Reduction on SPD Manifolds: The Emergence of Geometry-Aware Methods (TPAMI 2017 · similarity 0.86)
- Text-Driven Fusion for Infrared and Visible Images: Achieving Image Scene Adaptation on Hyperbolic Space (ICML 2026 · similarity 0.85)
- Equivariant Multi-Modality Image Fusion (CVPR 2024 · similarity 0.85)
- Revisiting Generative Infrared and Visible Image Fusion Based on Human Cognitive Laws (NeurIPS 2025 · similarity 0.85)
- Every SAM Drop Counts: Embracing Semantic Priors for Multi-Modality Image Fusion and Beyond (CVPR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
