DSFormer: A Dual-Scale Cross-Learning Transformer for Visual Place Recognition

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Jiang, Haiyang, Piao, Songhao, Gao, Chao, Yu, Lei, Chen, Liguo
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912499855523840
author Jiang, Haiyang
Piao, Songhao
Gao, Chao
Yu, Lei
Chen, Liguo
author_facet Jiang, Haiyang
Piao, Songhao
Gao, Chao
Yu, Lei
Chen, Liguo
contents Visual Place Recognition (VPR) is crucial for robust mobile robot localization, yet it faces significant challenges in maintaining reliable performance under varying environmental conditions and viewpoints. To address this, we propose a novel framework that integrates Dual-Scale-Former (DSFormer), a Transformer-based cross-learning module, with an innovative block clustering strategy. DSFormer enhances feature representation by enabling bidirectional information transfer between dual-scale features extracted from the final two CNN layers, capturing both semantic richness and spatial details through self-attention for long-range dependencies within each scale and shared cross-attention for cross-scale learning. Complementing this, our block clustering strategy repartitions the widely used San Francisco eXtra Large (SF-XL) training dataset from multiple distinct perspectives, optimizing data organization to further bolster robustness against viewpoint variations. Together, these innovations not only yield a robust global embedding adaptable to environmental changes but also reduce the required training data volume by approximately 30\% compared to previous partitioning methods. Comprehensive experiments demonstrate that our approach achieves state-of-the-art performance across most benchmark datasets, surpassing advanced reranking methods like DELG, Patch-NetVLAD, TransVPR, and R2Former as a global retrieval solution using 512-dim global descriptors, while significantly improving computational efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2507_18444
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DSFormer: A Dual-Scale Cross-Learning Transformer for Visual Place Recognition
Jiang, Haiyang
Piao, Songhao
Gao, Chao
Yu, Lei
Chen, Liguo
Computer Vision and Pattern Recognition
Robotics
Visual Place Recognition (VPR) is crucial for robust mobile robot localization, yet it faces significant challenges in maintaining reliable performance under varying environmental conditions and viewpoints. To address this, we propose a novel framework that integrates Dual-Scale-Former (DSFormer), a Transformer-based cross-learning module, with an innovative block clustering strategy. DSFormer enhances feature representation by enabling bidirectional information transfer between dual-scale features extracted from the final two CNN layers, capturing both semantic richness and spatial details through self-attention for long-range dependencies within each scale and shared cross-attention for cross-scale learning. Complementing this, our block clustering strategy repartitions the widely used San Francisco eXtra Large (SF-XL) training dataset from multiple distinct perspectives, optimizing data organization to further bolster robustness against viewpoint variations. Together, these innovations not only yield a robust global embedding adaptable to environmental changes but also reduce the required training data volume by approximately 30\% compared to previous partitioning methods. Comprehensive experiments demonstrate that our approach achieves state-of-the-art performance across most benchmark datasets, surpassing advanced reranking methods like DELG, Patch-NetVLAD, TransVPR, and R2Former as a global retrieval solution using 512-dim global descriptors, while significantly improving computational efficiency.
title DSFormer: A Dual-Scale Cross-Learning Transformer for Visual Place Recognition
topic Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2507.18444