Research map / Image generation / Image restoration
Wide2Long: Learning Lens Compression and Perspective Adjustment for Wide-Angle to Telephoto Translation
Soumyadipta Banerjee, Jiaul H. Paik, Debashis Sen
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 layer-wise in-painting (ablation) | W2L · W2L test set | PSNR: 9.28 -> 19.87 | Improved |
| add layer-wise in-painting (ablation) | W2L · CMDP portraits | PSNR: 7.29 -> 20.38 | Improved |
| add layer-wise in-painting (ablation) | W2L · W2L test set | SSIM: 0.463 -> 0.768 | Improved |
| add learned dilation parameter lambda (ablation) | W2L · W2L test set | PSNR: 16.47 -> 19.87 | Improved |
| add learned dilation parameter lambda (ablation) | W2L · CMDP portraits | PSNR: 16.21 -> 20.38 | Improved |
| add inter-layer redundancy reduction (ablation) | W2L · W2L test set | PSNR: 17.58 -> 19.87 | Improved |
| add inter-layer redundancy reduction (ablation) | W2L · CMDP portraits | SSIM: 0.492 -> 0.716 | Improved |
| use I2P net with 4 outputs instead of separate single-output I2P instances (ablation) | W2L · W2L test set | SSIM: 0.768 -> 0.483 | Hurt |
| use VGG-16 with 4 outputs instead of the I2P network (ablation) | W2L · W2L test set | SSIM: 0.768 -> 0.714 | Hurt |
| use VGG-16 with 1 output instead of the I2P network (ablation) | W2L · W2L test set | LPIPS: 0.139 -> 0.296 | Hurt |
| use W2L instead of Dolly-Zoom generation (Shih) | wide-to-telephoto translation · W2L test set | SSIM: 0.735 -> 0.768 | Improved |
| use W2L instead of U-Net translation | wide-to-telephoto translation · W2L test set | PSNR: 6.29 -> 19.873 | Improved |
Closest papers by meaning
- Beyond Wide-Angle Images: Structure-to-Detail Video Portrait Correction via Unsupervised Spatiotemporal Adaptation (AAAI 2026 · similarity 0.84)
- High-Resolution and Few-shot View Synthesis from Asymmetric Dual-lens Inputs (ECCV 2024 · similarity 0.84)
- Lifting the Structural Morphing for Wide-Angle Images Rectification: Unified Content and Boundary Modeling (ICCV 2025 · similarity 0.84)
- From Perspective to Fisheye Depth Estimation and Open-Vocabulary Segmentation (ECCV 2026 · similarity 0.84)
- Rethinking the Paradigm of Content Constraints in Unpaired Image-to-Image Translation (AAAI 2024 · similarity 0.83)
- LookHere: Vision Transformers with Directed Attention Generalize and Extrapolate (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.
