Research map / Image generation / Super-resolution
SCSNet: An Efficient Paradigm for Learning Simultaneously Image Colorization and Super-resolution
Jiangning Zhang, Chao Xu, Jian Li, Yue Han, Yabiao Wang, Ying Tai, Yong Liu
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 |
|---|---|---|---|
| Replace the cascaded colorization and super-resolution pipeline with end-to-end SCSNet | Automatic SCS · ImageNet-C | FID: 26.501 -> 25.992 | Improved |
| Use SCSNet rather than ColTran | Referential SCS · ImageNet-C | FID: 15.860 -> 9.632 | Improved |
| Add the perceptual loss to content loss (ablation) | SCSNet · ImageNet-C, referential | FID: 16.290 -> 15.068 | Improved |
| Add the adversarial loss to content loss (ablation) | SCSNet · ImageNet-C, referential | FID: 16.290 -> 11.616 | Improved |
| Use all three losses rather than content and perceptual losses only (ablation) | SCSNet · ImageNet-C, referential | FID: 15.068 -> 9.632 | Improved |
| Use all three losses rather than content and adversarial losses only (ablation) | SCSNet · ImageNet-C, referential | PSNR: 26.536 -> 27.694 | Improved |
| Replace basic cross-attention with PVCAttn (ablation) | SCSNet without CPM · ImageNet-C, referential | FID: 16.635 -> 15.334 | Improved |
| Add PVCAttn to the CPM variant (ablation) | SCSNet · ImageNet-C, referential | FID: 9.978 -> 9.632 | Improved |
| Add CPM to the PVCAttn variant (ablation) | SCSNet · ImageNet-C, referential | FID: 15.334 -> 9.632 | Improved |
| Use PVCAttn instead of BCAttn (ablation) | SCSNet without CPM · ImageNet-C, referential | CN: 4.767 -> 4.863 | Improved |
| Add CPM to the PVCAttn variant (ablation) | SCSNet · ImageNet-C, referential | CN: 4.863 -> 5.288 | Improved |
| Add PVCAttn to the CPM variant (ablation) | SCSNet · ImageNet-C, referential | CN: 5.059 -> 5.288 | Improved |
Closest papers by meaning
- Learning Texture Transformer Network for Image Super-Resolution (CVPR 2020 · similarity 0.87)
- Towards Real Scene Super-Resolution With Raw Images (CVPR 2019 · similarity 0.87)
- LGA-Net: Learning Local and Global Affinities for Sparse Scribble based Image Colorization (ICCV 2025 · similarity 0.87)
- ClassSR: A General Framework to Accelerate Super-Resolution Networks by Data Characteristic (CVPR 2021 · similarity 0.87)
- SeD: Semantic-Aware Discriminator for Image Super-Resolution (CVPR 2024 · similarity 0.86)
- Exploring Semantic Feature Discrimination for Perceptual Image Super-Resolution and Opinion-Unaware No-Reference Image Quality Assessment (CVPR 2025 · 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.
