Skip to content
Preprint

POI-Loc: A Fine-Grained POI Localization Benchmark and an Asymmetric Global-to-Local Matching Method

Sep 2026 · 0 citations · 25 references
Computer Science

TL;DR

This work introduces POI-Loc, to its knowledge the first benchmark dedicated to this asymmetric, fine-grained POI localization task, and proposes GLAM (Global-to-Local Asymmetric Matching), which surpasses both global and two-stage baselines on Recall@1/5/10 and mAP.

Abstract

Point-of-interest (POI) localization matches user-provided storefront close-ups to the same shops in wide, geo-tagged vehicle-mounted street views. POIs may change while the surrounding scene stays similar, so scene-level recognition alone cannot establish POI identity. Differences in target scale and capture domains further challenge matching. We introduce POI-Loc, to our knowledge the first benchmark dedicated to this asymmetric, fine-grained POI localization task. Many visual place recognition methods represent each image with a single global vector, which tends to dilute fine-grained features of small storefronts amid background clutter. We propose GLAM (Global-to-Local Asymmetric Matching) to combine global and local evidence. In stage one, a single attention-pooled query probe is matched against compact reference region tokens via learnable soft top-k interaction, with the resulting local similarity fused with global similarity for retrieval. Stage two reuses query region tokens before attention pooling and stored reference tokens for mutual-nearest-neighbor re-ranking. GLAM surpasses both global and two-stage baselines on Recall@1/5/10 and mAP, with about $5\times$ smaller re-ranking features and $280\times$ lower per-pair matching cost than FoL. The benchmark and code will be released at https://github.com/roadhan/glam.

View source

Similar papers

Preprint Aug 2026

MulVec: Fine-Grained Role-Aware Matching for Training-Free Zero-Shot Composed Image Retrieval

M UL V EC is proposed, a role-aware method whose compiler produces a structured query record that is mapped to four retrieval roles: Global describes the full target, Desired states what should appear, Preserve states what should remain, and Forbidden states what should disappear.

Zihao Zhang, Da-Yan Wu, Xin-Ze Liu et al. · 0 citations
#artificial intelligence Preprint Oct 2026

PARC-Loc: Text-to-Point-Cloud Localization with Partial Assignment and Relational Consistency

Text-to-point-cloud localization estimates a position in a city-scale 3D map from descriptions of surrounding objects. Existing coarse-to-fine methods retrieve submaps using aggregate learned compatibility and then localize within a selected submap. However, repetitive or similar urban objects can inflate the embedding...

Sheng-Kai Ma, Zhen-Yu Hou, Wei-Hua Cao · 0 citations
Preprint Aug 2026

MulVec: Fine-Grained Role-Aware Matching for Training-Free Zero-Shot Composed Image Retrieval

MULVEC is proposed, a role-aware method whose compiler produces a structured query record that is mapped to four retrieval roles: Global describes the full target, Desired states what should appear, Preserve states what should remain, and Forbidden states what should disappear.

Zihao Zhang, Da-Yan Wu, Xin-Ze Liu et al. · 0 citations
Preprint Aug 2026

UniGeo: A Multi-modal Large Language Model for Text-Guided Cross-View Geo-Localization

UniGeo, a unified multimodal large language model (MLLM) for text-guided drone geo-localization, is proposed and a multi-stage training strategy that progressively learns geo-semantic understanding, cross-view generation, and candidate verification, improving adaptation to text-guided geo-localization is introduced.

Jiahao Wen, Hang Yu, Zhedong Zheng · 0 citations
Oct 2026

Local Feature Enhancement of dino.txt for Training-Free Open-Vocabulary Semantic Segmentation

Open-vocabulary semantic segmentation aims to establish dense pixel-wise correspondences between images and open-ended semantic categories. While most existing methods build upon CLIP’s image-text alignment capability, they require substantial architectural modifications to adapt image-level understanding to pixel-leve...

Cheng He, Shi-Yu Xuan, Ze-Chao Li · 0 citations
2026

RAT-CVGL: Rank-Aware Transformer for Cross-View Geo-Localization

Cross-view geo-localization (CVGL) is a critical task that determines the geographic position of a query image via retrieving its corresponding counterparts across heterogeneous visual domains, such as ground-level, drone, and satellite views. Despite significant progress, recent existing approaches primarily focus on...

Guan-Bo Wang, Xu-Lei Shi, Xin Wang et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.