Research map / Label-efficient and robust learning / Self-supervised and contrastive
TransMEF: A Transformer-Based Multi-Exposure Image Fusion Framework Using Self-Supervised Multi-Task Learning
Linhao Qu, Shaolei Liu, Manning Wang, Zhijian Song
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 |
|---|---|---|---|
| add TransBlock (CNN+transformer encoder) to the encoder, with self-supervised tasks (ablation) | TransMEF · MEF benchmark (100 pairs), 20% training data | fusion objective metrics | Improved |
| add TransBlock to the encoder, without self-supervised tasks (ablation) | TransMEF · MEF benchmark (100 pairs), 20% training data | fusion objective metrics | Improved |
| use each single self-supervised reconstruction task alone (Gamma, Fourier, or region shuffling) (ablation) | TransMEF · MEF benchmark, 20% training data | fusion objective metrics | Improved |
| conduct all three self-supervised tasks simultaneously via multi-task learning instead of individually (ablation) | TransMEF · MEF benchmark, 20% training data | fusion objective metrics | Improved |
| use TransBlock encoder instead of CNN-only encoder for image reconstruction (ablation) | TransMEF encoder · image reconstruction examples | MSE, SSIM, PSNR | Improved |
| TransMEF vs 11 traditional and deep MEF methods | TransMEF · MEF benchmark, 12 objective metrics | 12 fusion metrics: best on nine of 12 metrics | Improved |
Closest papers by meaning
- Hybrid-Supervised Dual-Search: Leveraging Automatic Learning for Loss-Free Multi-Exposure Image Fusion (AAAI 2024 · similarity 0.87)
- Transformer Fusion with Optimal Transport (ICLR 2024 · similarity 0.84)
- Task-driven Image Fusion with Learnable Fusion Loss (CVPR 2025 · similarity 0.84)
- Task-Gated Multi-Expert Collaboration Network for Degraded Multi-Modal Image Fusion (ICML 2025 · similarity 0.84)
- Efficient Rectified Flow for Image Fusion (NeurIPS 2025 · similarity 0.84)
- Transforming Vision Transformer: Towards Efficient Multi-Task Asynchronous Learner (NeurIPS 2024 · similarity 0.83)
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.
