Research map / Label-efficient and robust learning / Recognition and long-tail
EfficientVPR: Toward Efficient Visual Place Recognition via Scene-Aware Prompt Tuning and Adaptive Feature Enhancement
Wenjing Tang, Chuanguang Yang, Zhulin An, Libo Huang, Boyu Diao, Yongjun Xu
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 |
|---|---|---|---|
| replace full fine-tuning with SceneVPT (ablation) | DINOv2-S · AmsterTime | R@1: 36.8 -> 48.4 | Improved |
| add SceneVPT adaptive prompts instead of VPT-deep (ablation) | DINOv2-S · AmsterTime | R@1: 48.2 -> 48.4 | Improved |
| use SceneVPT instead of adapter (ablation) | DINOv2-S · Pitts250k-test | R@1: 94.1 -> 94.5 | Improved |
| deep prompts across all layers instead of shallow (ablation) | VPT · Pitts250k-test | R@1: 93.0 -> 93.9 | Improved |
| use adaptive weighted prompt selection instead of random (ablation) | SceneVPT · AmsterTime | R@1: 52.1 -> 55.2 | Improved |
| use adaptive weighting instead of top-k selection (ablation) | SceneVPT · AmsterTime | R@1: 54.5 -> 55.2 | Improved |
| add OLE module (ablation) | EfficientVPR feature enhancement · AmsterTime | R@1: 47.1 -> 48.0 | Improved |
| add MsIA module (ablation) | EfficientVPR feature enhancement · AmsterTime | R@1: 48.0 -> 48.4 | Improved |
| replace linear projection with SGFS (ablation) | EfficientVPR · AmsterTime | R@1: 48.4 -> 55.2 | Improved |
| use SGFS instead of BoQ block (ablation) | EfficientVPR · AmsterTime | R@1: 51.1 -> 55.2 | Improved |
| increase inference image size from 322 to 504 (ablation) | EfficientVPR · AmsterTime | R@1: 60.2 -> 61.4 | Mixed |
| train at larger image size 280 instead of 266 (ablation) | EfficientVPR · Pitts250k-test | R@1: 95.4 -> 95.2 | Mixed |
Closest papers by meaning
- EffoVPR: Effective Foundation Model Utilization for Visual Place Recognition (ICLR 2025 · similarity 0.92)
- Towards Seamless Adaptation of Pre-trained Models for Visual Place Recognition (ICLR 2024 · similarity 0.90)
- MutualVPR: A Mutual Learning Framework for Resolving Supervision Inconsistencies via Adaptive Clustering (NeurIPS 2025 · similarity 0.90)
- Focus on Local: Finding Reliable Discriminative Regions for Visual Place Recognition (AAAI 2025 · similarity 0.89)
- Optimal Transport Aggregation for Visual Place Recognition (CVPR 2024 · similarity 0.89)
- CricaVPR: Cross-image Correlation-aware Representation Learning for Visual Place Recognition (CVPR 2024 · similarity 0.89)
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.
