Evaluation of Visual Place Recognition Methods for Image Pair Retrieval in 3D Vision and Robotics

2026-03-14

AI Summary

This work reframes visual place recognition (VPR) as an image-pair retrieval task to better support downstream applications such as scene registration, SLAM, and structure-from-motion. The study systematically evaluates prominent VPR methods—including NetVLAD, CosPlace, EigenPlaces, MixVPR, AnyLoc, SALAD, and MegaLoc—on three cross-domain 3D datasets: Tanks and Temples, ScanNet-GS, and KITTI. For the first time, VPR is explicitly modeled as a front-end for image-pair retrieval, revealing the domain dependency of existing approaches under challenges like perceptual aliasing and sequence incompleteness. Experimental results demonstrate that modern global descriptors can serve as plug-and-play, efficient retrieval modules, offering practical guidance for robust 3D mapping and registration pipelines.

Technology Category

• Computer Vision: Image and Video Retrieval
• Intelligent Robots: Localization, Mapping, and Navigation
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• 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

Visual Place Recognition (VPR) is a core component in computer vision, typically formulated as an image retrieval task for localization, mapping, and navigation. In this work, we instead study VPR as an image pair retrieval front-end for registration pipelines, where the goal is to find top-matching image pairs between two disjoint image sets for downstream tasks such as scene registration, SLAM, and Structure-from-Motion. We comparatively evaluate state-of-the-art VPR families - NetVLAD-style baselines, classification-based global descriptors (CosPlace, EigenPlaces), feature-mixing (MixVPR), and foundation-model-driven methods (AnyLoc, SALAD, MegaLoc) - on three challenging datasets: object-centric outdoor scenes (Tanks and Temples), indoor RGB-D scans (ScanNet-GS), and autonomous-driving sequences (KITTI). We show that modern global descriptor approaches are increasingly suitable as off-the-shelf image pair retrieval modules in challenging scenarios including perceptual aliasing and incomplete sequences, while exhibiting clear, domain-dependent strengths and weaknesses that are critical when choosing VPR components for robust mapping and registration.

Problem

• Visual Place Recognition
• Image Pair Retrieval
• 3D Vision
• Robotics
• Scene Registration

Innovation

• Visual Place Recognition
• Image Pair Retrieval
• Global Descriptors
• 3D Registration
• Foundation Models