AI Paper Atlas

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

CVPR 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning

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 changedWhereMeasuredEffect
replace full fine-tuning with SceneVPT (ablation)DINOv2-S · AmsterTimeR@1: 36.8 -> 48.4Improved
add SceneVPT adaptive prompts instead of VPT-deep (ablation)DINOv2-S · AmsterTimeR@1: 48.2 -> 48.4Improved
use SceneVPT instead of adapter (ablation)DINOv2-S · Pitts250k-testR@1: 94.1 -> 94.5Improved
deep prompts across all layers instead of shallow (ablation)VPT · Pitts250k-testR@1: 93.0 -> 93.9Improved
use adaptive weighted prompt selection instead of random (ablation)SceneVPT · AmsterTimeR@1: 52.1 -> 55.2Improved
use adaptive weighting instead of top-k selection (ablation)SceneVPT · AmsterTimeR@1: 54.5 -> 55.2Improved
add OLE module (ablation)EfficientVPR feature enhancement · AmsterTimeR@1: 47.1 -> 48.0Improved
add MsIA module (ablation)EfficientVPR feature enhancement · AmsterTimeR@1: 48.0 -> 48.4Improved
replace linear projection with SGFS (ablation)EfficientVPR · AmsterTimeR@1: 48.4 -> 55.2Improved
use SGFS instead of BoQ block (ablation)EfficientVPR · AmsterTimeR@1: 51.1 -> 55.2Improved
increase inference image size from 322 to 504 (ablation)EfficientVPR · AmsterTimeR@1: 60.2 -> 61.4Mixed
train at larger image size 280 instead of 266 (ablation)EfficientVPR · Pitts250k-testR@1: 95.4 -> 95.2Mixed

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy