Research map / Image generation / Image restoration
Lifting the Structural Morphing for Wide-Angle Images Rectification: Unified Content and Boundary Modeling
Wenting Luan, Siqi Lu, Yongbin Zheng, Wanying Xu, Lang Nie, Zongtan Zhou, Kang Liao
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 structural morphing (ablation) | ConBo-Net baseline · Wide-angle rectification+rectangling test (500 pairs) | PSNR: 18.1041 -> 18.5560 | Improved |
| add structural morphing (ablation) | ConBo-Net baseline · Wide-angle rectification+rectangling test (500 pairs) | LPIPS: 0.1563 -> 0.1284 | Improved |
| add structural morphing (ablation) | ConBo-Net baseline · Wide-angle rectification+rectangling test (500 pairs) | SSIM: 0.5271 -> 0.5366 | Improved |
| add ordinal geometry loss (ablation) | ConBo-Net baseline · Wide-angle rectification+rectangling test (500 pairs) | PSNR: 18.1041 -> 18.1743 | Improved |
| add ordinal geometry loss (ablation) | ConBo-Net baseline · Wide-angle rectification+rectangling test (500 pairs) | LPIPS: 0.1563 -> 0.1523 | Improved |
| add ordinal geometry loss (ablation) | ConBo-Net with morphing · Wide-angle rectification+rectangling test (500 pairs) | PSNR: 18.5560 -> 18.5922 | Improved |
| add ordinal geometry loss (ablation) | ConBo-Net with morphing · Wide-angle rectification+rectangling test (500 pairs) | LPIPS: 0.1284 -> 0.1300 | Hurt |
| add ordinal geometry loss (ablation) | ConBo-Net with morphing · Wide-angle rectification+rectangling test (500 pairs) | MS-SSIM: 0.7485 -> 0.7492 | Improved |
| one-stage unified model instead of two-stage cascade (DR-GAN+RecRecNet) | ConBo-Net · Wide-angle rectification+rectangling test (500 pairs) | PSNR: 18.09 -> 18.59 | Improved |
| one-stage unified model instead of two-stage cascade (DR-GAN+RecRecNet) | ConBo-Net · Wide-angle rectification+rectangling test (500 pairs) | mAP (YOLO): 0.24 -> 0.46 | Improved |
| one-stage unified model instead of ODE+RecRecNet | ConBo-Net · Wide-angle rectification+rectangling test (500 pairs) | Time (s): 0.02 -> 0.04 | Hurt |
| one-stage instead of rectification then He et al. rectangling | ConBo-Net · Wide-angle rectification+rectangling test (500 pairs) | Time (s): 19.48 -> 0.04 | Improved |
Closest papers by meaning
- RecRecNet: Rectangling Rectified Wide-Angle Images by Thin-Plate Spline Model and DoF-based Curriculum Learning (ICCV 2023 · similarity 0.91)
- Towards Complete Scene and Regular Shape for Distortion Rectification by Curve-Aware Extrapolation (ICCV 2021 · similarity 0.89)
- Wide-Angle Image Rectification: A Survey (IJCV 2022 · similarity 0.89)
- FLOWING: Implicit Neural Flows for Structure-Preserving Morphing (NeurIPS 2025 · similarity 0.86)
- RecDiffusion: Rectangling for Image Stitching with Diffusion Models (CVPR 2024 · similarity 0.86)
- MorpheuS: Neural Dynamic 360deg Surface Reconstruction from Monocular RGB-D Video (CVPR 2024 · similarity 0.85)
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.
