Research map / Label-efficient and robust learning / Recognition and long-tail
Progressive Large Scale-Invariant Image Matching in Scale Space
Lei Zhou, Siyu Zhu, Tianwei Shen, Jinglu Wang, Tian Fang, Long Quan
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 scale-invariant similarity metric | vocabulary-tree retrieval · Oxford | mAP: 0.515 -> 0.562 | Improved |
| add scale-invariant similarity metric | vocabulary-tree retrieval · Holidays | mAP: 0.607 -> 0.646 | Improved |
| add scale-invariant similarity metric | vocabulary-tree retrieval · Oxford | retrieval time (s): 840.9 -> 859.6 | Hurt |
| add scale-invariant similarity metric | baseline retrieval with no extra technique · Multi-scale retrieval set | mAP: 0.397 -> 0.672 | Improved |
| add scale-invariant similarity metric on top of WGC | baseline retrieval · Multi-scale retrieval set | mAP: 0.440 -> 0.679 | Improved |
| add scale-invariant similarity metric on top of PGM+HE+MA | baseline retrieval · Multi-scale retrieval set | mAP: 0.554 -> 0.734 | Improved |
| use scale-invariant matcher instead of exhaustive SIFT matcher | SIFT matching · Mikolajczyk zoom+rotation, pair 5 | correct matches: 112 -> 178 | Improved |
| use scale-invariant matcher instead of exhaustive SIFT matcher | SIFT matching · Mikolajczyk zoom+rotation | average matching time (ms): 59.16 -> 33.26 | Improved |
| use scale-invariant matcher instead of exhaustive SIFT matcher | SIFT matching · Mikolajczyk viewpoint, pair 3 | correct matches: 64 -> 91 | Improved |
| use scale-invariant matcher under large viewpoint change | SIFT matching · Mikolajczyk viewpoint, pairs 4 and 5 | correct matches: 0 -> 0 | No clear effect |
| use scale-invariant matcher instead of MODS | feature matching · Book-cover set, scale ratio 55 | feature matches: 0 -> 13 | Improved |
| use scale-invariant matcher instead of KVLD | feature matching · Book-cover set, scale ratio 32 | feature matches: 44 -> 60 | Improved |
Closest papers by meaning
- SIFTing Through Scales (TPAMI 2016 · similarity 0.91)
- Dense Correspondences across Scenes and Scales (TPAMI 2016 · similarity 0.89)
- Raising the Ceiling: Conflict-Free Local Feature Matching with Dynamic View Switching (ECCV 2024 · similarity 0.87)
- Learning Dense Feature Matching via Lifting Single 2D Image to 3D Space (ICCV 2025 · similarity 0.85)
- ScaleNet: A Shallow Architecture for Scale Estimation (CVPR 2022 · similarity 0.85)
- MATCHA: Towards Matching Anything (CVPR 2025 · 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.
