Research map / Vision-language and multimodal / Cross-modal learning
Deep Convolutional Neural Network for Multi-Modal Image Restoration and Fusion
Xin Deng, Pier Luigi Dragotti
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 residual connections in UFEM and CFPM (ablation) | CU-Net · RGB/depth SR 4x | PSNR (dB): 43.22 -> 44.65 (RMSE 2.03 -> 1.88) | Improved |
| increase filter size from 3 to 8 (ablation) | CU-Net · RGB/depth SR 4x | PSNR (dB): 43.45 -> 44.65 | Improved |
| increase filter size from 8 to 9 (ablation) | CU-Net · RGB/depth SR 4x | PSNR (dB): 44.65 -> 44.44 | Hurt |
| increase number of LCSC blocks per module from 1 to 6 (ablation) | CU-Net · RGB/depth SR 4x | PSNR (dB): 43.88 -> 45.12 (model size 66 KB -> 231 KB) | Improved |
| increase LCSC blocks from 4 to 6 (ablation) | CU-Net · RGB/depth SR 4x | RMSE: 1.88 -> 1.81 (165 KB -> 231 KB) | Improved |
| use CU-Net instead of DJFR (ablation) | CU-Net · RGB/MS SR 4x | PSNR: 39.28 -> 40.83 | Improved |
| use CU-Net instead of CoISTA | CU-Net · RGB/depth SR 4x | RMSE: 2.01 -> 1.88 | Improved |
| use CU-Net instead of DnCNN | CU-Net · Flash/non-flash denoising (sigma^2=75) | PSNR: 30.08 -> 31.30 | Improved |
| use CU-Net instead of DnCNN | CU-Net · Flash/non-flash denoising (sigma^2=25) | PSNR: 35.38 -> 35.86 | Improved |
| add common reconstruction to unique reconstruction (ablation) | CU-Net · RGB guided depth SR | visual quality | Improved |
Closest papers by meaning
- Revitalizing Convolutional Network for Image Restoration (TPAMI 2024 · similarity 0.88)
- Learning Enriched Features for Real Image Restoration and Enhancement (ECCV 2020 · similarity 0.88)
- Learning Enriched Features for Fast Image Restoration and Enhancement (TPAMI 2022 · similarity 0.88)
- Neural Sparse Representation for Image Restoration (NeurIPS 2020 · similarity 0.87)
- Deep M2CDL: Deep Multi-Scale Multi-Modal Convolutional Dictionary Learning Network (TPAMI 2023 · similarity 0.87)
- Omni-Kernel Network for Image Restoration (AAAI 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.
