Research map / Image generation / Super-resolution
Learning To Zoom Inside Camera Imaging Pipeline
Chengzhou Tang, Yuqiang Yang, Bing Zeng, Ping Tan, Shuaicheng Liu
Check 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 RAW-to-RAW super-resolution with jointly estimated image and degradation subspaces | ZIC · RealSR ×4 | PSNR: 27.89 -> 32.58 | Improved |
| Use RAW-to-RAW super-resolution with jointly estimated image and degradation subspaces | ZIC · SR-RAW ×4 | PSNR: 27.56 -> 32.89 | Improved |
| Add a spatially varying kernel map instead of using a global kernel (ablation) | ZIC · RealSR ×4 | PSNR: 31.63 -> 32.58 | Improved |
| Add a spatially varying kernel map instead of using a global kernel (ablation) | ZIC · SR-RAW ×4 | PSNR: 32.24 -> 32.89 | Improved |
| Perform super-resolution in RAW-to-RAW rather than RGB-to-RGB (ablation) | ZIC · RealSR ×4 | PSNR: 31.91 -> 32.58 | Improved |
| Perform super-resolution in RAW-to-RAW rather than RGB-to-RGB (ablation) | ZIC · SR-RAW ×4 | PSNR: 32.38 -> 32.89 | Improved |
| Jointly minimize image and kernel coefficients rather than alternating their updates (ablation) | ZIC · RealSR ×4 | PSNR: 32.10 -> 32.58 | Improved |
| Jointly minimize image and kernel coefficients rather than alternating their updates (ablation) | ZIC · SR-RAW ×4 | PSNR: 32.65 -> 32.89 | Improved |
| Use 4 kernel-subspace dimensions and 8 image-subspace dimensions (ablation) | ZIC · RealSR ×4 | SSIM | Improved |
| Train on SR-RAW and transfer to RealSR without fine-tuning | ZIC · RealSR ×4, zero-shot transfer | PSNR: 26.57 -> 32.07 | Improved |
Closest papers by meaning
- Towards Real Scene Super-Resolution With Raw Images (CVPR 2019 · similarity 0.91)
- Exploiting Raw Images for Real-Scene Super-Resolution (TPAMI 2020 · similarity 0.90)
- Zoom to Learn, Learn to Zoom (CVPR 2019 · similarity 0.90)
- Toward Real-World Single Image Super-Resolution: A New Benchmark and a New Model (ICCV 2019 · similarity 0.90)
- Learning a Single Convolutional Super-Resolution Network for Multiple Degradations (CVPR 2018 · similarity 0.89)
- Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network (CVPR 2016 · similarity 0.89)
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.
