Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Match-Any-Events: Zero-Shot Motion-Robust Feature Matching Across Wide Baselines for Event Cameras
Ruijun Zhang, Hang Su, Kostas Daniilidis, Ziyun Wang
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 Temporal Aggregation Transformer (TAg) (ablation) | Match-Any-Events · ECM | AUC@5: 50.12 -> 54.61 | Improved |
| add Temporal Aggregation Transformer (TAg) (ablation) | Match-Any-Events · ECM | Precision (%): 67.18 -> 68.90 | Improved |
| use TAg instead of Concentrate Net (ablation) | Match-Any-Events · ECM | AUC@5: 48.37 -> 54.61 | Improved |
| add multi-scale input (ablation) | Match-Any-Events w/o TAg · ECM | AUC@5: 39.86 -> 50.12 | Improved |
| add synthetic E-MegaDepth data to real-only training (ablation) | Match-Any-Events · ECM | AUC@5: 11.33 -> 54.61 | Improved |
| add real M3ED data to synthetic-only training (ablation) | Match-Any-Events · ECM | AUC@5: 52.63 -> 54.61 | Improved |
| add real M3ED data to synthetic-only training (ablation) | Match-Any-Events · ECM | Precision (%): 68.04 -> 68.90 | Improved |
| use Match-Any-Events instead of SuperEvent | SuperEvent · EDS pose estimation | AUC@5: 25.4 -> 40.4 | Improved |
| feed intensity-reconstructed video (E2V) instead of raw event frames to MatchAnything | MatchAnything · ECM event-to-event | AUC@5: 20.69 -> 41.68 | Improved |
| feed E2V reconstruction instead of event frames to VGGT | VGGT · ECM event-to-event | AUC@5: 1.86 -> 31.44 | Improved |
Closest papers by meaning
- ETAP: Event-based Tracking of Any Point (CVPR 2025 · similarity 0.89)
- MATE: Motion-Augmented Temporal Consistency for Event-based Point Tracking (ICCV 2025 · similarity 0.89)
- Data-Driven Feature Tracking for Event Cameras (CVPR 2023 · similarity 0.88)
- Data Association between Event Streams and Intensity Frames under Diverse Baselines (ECCV 2022 · similarity 0.88)
- PASS: Path-selective State Space Model for Event-based Recognition (NeurIPS 2025 · similarity 0.87)
- FlexEvent: Towards Flexible Event-Frame Object Detection at Varying Operational Frequencies (NeurIPS 2025 · similarity 0.87)
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.
