Focus on Local: Finding Reliable Discriminative Regions for Visual Place Recognition

2025-04-11

AAAI Conference on Artificial Intelligence

AI Summary

To address insufficient discriminative local region modeling, background-induced false matches, and the lack of local supervision—leading to compromised efficiency and accuracy in visual place recognition (VPR)—this paper proposes a weakly supervised local feature learning framework. Our method introduces: (1) a novel spatial alignment loss (SAL) and foreground-background contrastive enhancement loss (CEL) to enable precise localization and discriminability enhancement of local regions; (2) a weakly supervised local training paradigm that generates pseudo-correspondences from global features, eliminating the need for manual annotations; and (3) an efficient, discriminative-region-guided re-ranking pipeline. Evaluated on mainstream VPR benchmarks, our approach achieves state-of-the-art performance in both image retrieval and re-ranking. Compared to two-stage methods, it improves re-ranking speed by 3.2× and recall by 4.7%.

Technology Category

• Computer Vision: Image and Video Retrieval
• Search and Optimization: Local Search
• Intelligent Robots: Localization, Mapping, and Navigation

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Visual Place Recognition (VPR) is aimed at predicting the location of a query image by referencing a database of geotagged images. For VPR task, often fewer discriminative local regions in an image produce important effects while mundane background regions do not contribute or even cause perceptual aliasing because of easy overlap. However, existing methods lack precisely modeling and full exploitation of these discriminative regions. In addition, the lack of pixel-level correspondence supervision in the VPR dataset hinders further improvement of the local feature matching capability in the re-ranking stage. In this paper, we propose the Focus on Local (FoL) approach to stimulate the performance of image retrieval and re-ranking in VPR simultaneously by mining and exploiting reliable discriminative local regions in images and introducing pseudo-correlation supervision. First, we design two losses, Extraction-Aggregation Spatial Alignment Loss (SAL) and Foreground-Background Contrast Enhancement Loss (CEL), to explicitly model reliable discriminative local regions and use them to guide the generation of global representations and efficient re-ranking. Second, we introduce a weakly-supervised local feature training strategy based on pseudo-correspondences obtained from aggregating global features to alleviate the lack of local correspondences ground truth for the VPR task. Third, we suggest an efficient re-ranking pipeline that is efficiently and precisely based on discriminative region guidance. Finally, experimental results show that our FoL achieves the state-of-the-art on multiple VPR benchmarks in both image retrieval and re-ranking stages and also significantly outperforms existing two-stage VPR methods in terms of computational efficiency.

Problem

• Identifying reliable discriminative regions for Visual Place Recognition
• Lack of precise modeling in existing VPR methods
• Improving image retrieval and re-ranking efficiency in VPR

Innovation

• Mining reliable discriminative local regions
• Weakly-supervised local feature training strategy
• Efficient re-ranking pipeline with region guidance