Research map / 3D vision / Depth and stereo
Beyond Tie Points: Satellite Image Block Adjustment based on Dense Feature Consistency
Yi Liu, Yi Wan, Lei Yu, Panwang Xia, Qiong Wu, Yingying Pei, Xuejun Huang, Junjian Zhang, Xiangyuan Cai, Hongwei Hu, Yongjun Zhang
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 geographic-coordinate-regression adapter to frozen features (ablation) | Ours · Guangzhou-b (ablation) | mean error (m): 5.57 -> 4.66 | Improved |
| add confidence weighting without adapter (ablation) | Ours · Guangzhou-b (ablation) | mean error (m): 5.57 -> 5.55 | No clear effect |
| add confidence weighting on top of adapter (ablation) | Ours · Guangzhou-b (ablation) | mean error (m): 4.66 -> 1.84 | Improved |
| add confidence weighting on top of adapter (ablation) | Ours · Guangzhou-b (ablation) | percentage within 3 m: 18.33 -> 88.33 | Improved |
| use dense feature consistency instead of tie points (vs PBA-LoFTR) | Ours · Beijing small initial error | mean error (m): 5.87 -> 2.24 | Improved |
| use dense feature consistency instead of tie points (vs PBA-LoFTR) | Ours · Guangzhou small initial error | mean error (m): 5.06 -> 1.42 | Improved |
| add confidence filtering to LoFTR tie points | PBA-LoFTR · Beijing small initial error | mean error (m): 5.87 -> 4.03 | Improved |
| use full method instead of LoFTR with confidence filtering | Ours · Beijing small initial error | mean error (m): 4.03 -> 2.24 | Improved |
| use full method instead of LoFTR with confidence filtering | Ours · San Jose large initial error | mean error (m): 0.56 -> 0.75 | Hurt |
| use full method on San Jose small initial error | Ours · San Jose small initial error | mean error (m): 0.53 -> 0.50 | Improved |
Closest papers by meaning
- RSL-BA: Rolling Shutter Line Bundle Adjustment (ECCV 2024 · similarity 0.81)
- SkySplat: Generalizable 3D Gaussian Splatting from Multi-Temporal Sparse Satellite Images (AAAI 2026 · similarity 0.81)
- SIPSA-Net: Shift-Invariant Pan Sharpening With Moving Object Alignment for Satellite Imagery (CVPR 2021 · similarity 0.80)
- Accurate 3-DoF Camera Geo-Localization via Ground-to-Satellite Image Matching (TPAMI 2022 · similarity 0.80)
- Efficient Second-Order Plane Adjustment (CVPR 2023 · similarity 0.80)
- Satellite to GroundScape - Large-scale Consistent Ground View Generation from Satellite Views (CVPR 2025 · similarity 0.80)
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.
