Research map / Image generation / Image restoration
Weakly-Supervised Stitching Network for Real-World Panoramic Image Generation
Dae-Young Song, Geonsoo Lee, HeeKyung Lee, Gi-Mun Um, Donghyeon Cho
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 post-color correction map (ablation) | Weakly-supervised stitching network · Own test set | P_d: 2.773 -> 2.731 | Improved |
| add post-color correction map (ablation) | Weakly-supervised stitching network · Own test set | SIQE: 39.528 -> 37.714 | Hurt |
| add post-color correction map (ablation) | Weakly-supervised stitching network · Own test set | FID: 132.4 -> 140.8 | Hurt |
| use both pre- and post-color correction instead of either alone (ablation) | Weakly-supervised stitching network · Own test set (qualitative) | visual quality | Improved |
| use perceptual loss instead of pixel-wise L1 loss (ablation) | Weakly-supervised stitching network · Own test set (qualitative) | visual quality | Improved |
| increase number of warpings K from 1 to 4 (ablation) | Weakly-supervised stitching network · Own test set (14 sets) | P_d: 4.001 -> 3.904 | No clear effect |
| increase number of warpings K from 1 to 5 (ablation) | Weakly-supervised stitching network · Own test set (14 sets) | SIQE: 22.89 -> 18.29 | Hurt |
| use proposed model instead of WeiMethod | Weakly-supervised stitching network · CROSS | PSNR: 18.730 -> 22.440 | Improved |
| use GPU instead of CPU inference | Weakly-supervised stitching network · Own test set | time per image (s): 1.3870 -> 0.0363 | Improved |
Closest papers by meaning
- Image Stitching in Adverse Condition: A Bidirectional-Consistency Learning Framework and Benchmark (NeurIPS 2025 · similarity 0.89)
- Deep Rectangling for Image Stitching: A Learning Baseline (CVPR 2022 · similarity 0.88)
- Reconstructing the Image Stitching Pipeline: Integrating Fusion and Rectangling into a Unified Inpainting Model (NeurIPS 2024 · similarity 0.87)
- Depth-Supervised Fusion Network for Seamless-Free Image Stitching (NeurIPS 2025 · similarity 0.87)
- Towards Robust Image Stitching: An Adaptive Resistance Learning against Compatible Attacks (AAAI 2024 · similarity 0.85)
- Efficient Stitchable Task Adaptation (CVPR 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.
