Research map / Vision-language and multimodal / Cross-modal learning
MMMamba: A Versatile Cross-Modal in Context Fusion Framework for Pan-Sharpening and Zero-Shot Image Enhancement
Yingying Wang, Xuanhua He, Chen Wu, Jialing Huang, Suiyun Zhang, Rui Liu, Xinghao Ding, Haoxuan Che
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 Mamba instead of self-attention (ablation) | MMMamba · WorldView-II | PSNR: 41.3995 -> 42.3120 | Improved |
| use in-context fusion instead of channel concatenation (ablation) | MMMamba · WorldView-II | PSNR: 41.2898 -> 42.3120 | Improved |
| interleave PAN and MS tokens instead of sequential concatenation (ablation) | MMMamba · WorldView-II | PSNR: 36.4702 -> 42.3120 | Improved |
| use four-direction scanning instead of one-way scan (ablation) | MMMamba · WorldView-II | PSNR: 42.0965 -> 42.3120 | Improved |
| use local window scan instead of global scan (ablation) | MMMamba · WorldView-II | PSNR: 42.1998 -> 42.3120 | Improved |
| MMMamba vs best prior method on full-resolution GF2 metrics | MMMamba · GF2 full-resolution | D_lambda / D_S / QNR: D_lambda 0.0656 vs 0.0652 (Pan-Mamba); QNR 0.8312 vs 0.8471 (SFINet++) | Mixed |
| MMMamba vs Pan-Mamba on reduced-resolution WV3 | MMMamba · WorldView-III | ERGAS: 2.8950 vs 2.8910 | No clear effect |
| drop PAN input to do zero-shot super-resolution | MMMamba · WV2 zero-shot super-resolution | PSNR: 34.0869 (Bicubic) -> 36.4892 | Improved |
| use Mamba instead of Transformer on cost (ablation) | MMMamba · Efficiency | Params (M): 0.3684 -> 0.2453 params; 3.9206 -> 5.0616 GFLOPs | Mixed |
Closest papers by meaning
- Wavelet-Assisted Multi-Frequency Attention Network for Pansharpening (AAAI 2025 · similarity 0.87)
- Accelerated Diffusion via High-Low Frequency Decomposition for Pan-Sharpening (AAAI 2025 · similarity 0.87)
- Unsupervised Pan-Sharpening via Mutually Guided Detail Restoration (AAAI 2024 · similarity 0.87)
- Binarized Neural Network for Multi-spectral Image Fusion (CVPR 2025 · similarity 0.86)
- PanAdapter: Two-Stage Fine-Tuning with Spatial-Spectral Priors Injecting for Pansharpening (AAAI 2025 · similarity 0.86)
- GeminiFusion: Efficient Pixel-wise Multimodal Fusion for Vision Transformer (ICML 2024 · similarity 0.86)
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.
