AIR-HLoc: Adaptive Retrieved Images Selection for Efficient Visual Localisation

2024-03-27

AI Summary

In hierarchical visual localization, the image retrieval stage suffers from computational redundancy due to fixed top-k selection. Method: This paper proposes a global descriptor similarity–driven adaptive strategy for selecting the number of retrieved images *k*. We first empirically reveal a strong positive correlation between global similarity scores and local feature matching success rates, and accordingly design a dynamic *k*-selection mechanism that breaks the traditional linear computational growth bottleneck. The method integrates global retrieval (e.g., NetVLAD, CosPlace), local matching (SuperPoint + SuperGlue), and similarity-guided adaptive thresholding. Results: Evaluated on three large-scale indoor and outdoor benchmarks, our approach reduces feature matching latency by up to 30% while achieving state-of-the-art localization accuracy, significantly enhancing practicality in low-latency applications.

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
• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

State-of-the-art hierarchical localisation pipelines (HLoc) employ image retrieval (IR) to establish 2D-3D correspondences by selecting the top-$k$ most similar images from a reference database. While increasing $k$ improves localisation robustness, it also linearly increases computational cost and runtime, creating a significant bottleneck. This paper investigates the relationship between global and local descriptors, showing that greater similarity between the global descriptors of query and database images increases the proportion of feature matches. Low similarity queries significantly benefit from increasing $k$, while high similarity queries rapidly experience diminishing returns. Building on these observations, we propose an adaptive strategy that adjusts $k$ based on the similarity between the query's global descriptor and those in the database, effectively mitigating the feature-matching bottleneck. Our approach optimizes processing time without sacrificing accuracy. Experiments on three indoor and outdoor datasets show that AIR-HLoc reduces feature matching time by up to 30%, while preserving state-of-the-art accuracy. The results demonstrate that AIR-HLoc facilitates a latency-sensitive localisation system.

Problem

• Optimizes image retrieval for efficient visual localization.
• Reduces computational cost by adaptive image selection.
• Maintains accuracy while decreasing feature matching time.

Innovation

• Adaptive image retrieval strategy optimizes k.
• Reduces feature matching time by 30%.
• Maintains accuracy while improving computational efficiency.