Research map / Image generation / Image restoration
Leveraging Local Patch Alignment to Seam-cutting for Large Parallax Image Stitching
Tianli Liao, Chenyang Zhao, Lei Li, Heling Cao
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 LPAM | Baseline seam-cutting · Parallax (35 pairs, REW alignment) | SSIM: 0.750 -> 0.782 | Improved |
| add LPAM | Perception seam-cutting · Parallax (35 pairs, REW alignment) | SSIM: 0.749 -> 0.795 | Improved |
| add LPAM | Iterative seam-cutting · Parallax (35 pairs, REW alignment) | SSIM: 0.762 -> 0.800 | Improved |
| add LPAM | Quaternion seam-cutting · Parallax (35 pairs, REW alignment) | SSIM: 0.746 -> 0.783 | Improved |
| add LPAM | Baseline seam-cutting · SEAGULL (24 pairs, REW alignment) | PSNR: 26.20 -> 28.84 | Improved |
| add LPAM | Baseline seam-cutting (REW) · Our dataset (20 pairs) | PSNR: 28.50 -> 28.76 | Improved |
| add LPAM | UDIS++ alignment · Parallax (35 pairs, Baseline seam-cutting) | SSIM: 0.643 -> 0.705 | Improved |
| add LPAM | Homo alignment · Parallax (35 pairs, Baseline seam-cutting) | SSIM: 0.692 -> 0.725 | Improved |
| add LPAM | Iterative seam-cutting · Parallax (35 pairs) | elapsed time (s): 112.04 -> 8.85 | Improved |
| add LPAM | Baseline seam-cutting · Parallax (35 pairs) | elapsed time (s): 7.54 -> 7.77 | No clear effect |
| add sigmoid flow modification (beta=8) instead of raw SIFT flow (ablation) | LPAM · Parallax (35 pairs, Baseline seam-cutting) | SSIM: 0.819 -> 0.782 | Hurt |
| increase sigmoid beta from 1 to 16 (ablation) | LPAM · Parallax (35 pairs, Baseline seam-cutting) | SSIM: 0.770 -> 0.784 | Improved |
Closest papers by meaning
- SEAGULL: Seam-Guided Local Alignment for Parallax-Tolerant Image Stitching (ECCV 2016 · similarity 0.90)
- Leveraging Line-Point Consistence To Preserve Structures for Wide Parallax Image Stitching (CVPR 2021 · similarity 0.89)
- Warping Residual Based Image Stitching for Large Parallax (CVPR 2020 · similarity 0.89)
- Depth-Supervised Fusion Network for Seamless-Free Image Stitching (NeurIPS 2025 · similarity 0.88)
- Object-level Geometric Structure Preserving for Natural Image Stitching (AAAI 2025 · similarity 0.84)
- Reconstructing the Image Stitching Pipeline: Integrating Fusion and Rectangling into a Unified Inpainting Model (NeurIPS 2024 · similarity 0.82)
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.
