What type of study is this?

August 20, 2025Open Access

Visual Localisation Using Deep Learning and Graph Neural Networks: Approaches and Evaluation

Key Points

SuperGlue model achieved a mean absolute error of 6.37266, outperforming traditional methods.
Evaluation conducted on a dataset from the Getty Center in Los Angeles for a Kaggle competition.
Implementation involved models like SIFT, CNN-based baseline, and an ImageSimilarity-Autoencoder.
Integration of advanced architectures and attention mechanisms may improve localization tasks in real-world scenarios.

Abstract

This paper addresses the visual localization problem, focusing on estimating camera position and orientation from images in a known scene. Traditional localization methods utilizing local feature matching face challenges in generalization to new scenarios. In contrast, this study explores state-of-the-art techniques, including deep learning models and graph neural networks, to enhance feature extraction and matching. We implemented five models: SIFT, a CNN-based baseline, Hierarchical Localisation, an ImageSimilarity-Autoencoder, and the SuperGlue feature matching model. Evaluated on a dataset from the Getty Center in Los Angeles for a Kaggle competition, the SuperGlue model significantly outperformed others, achieving a mean absolute error (MAE) of 6.37266. The findings suggest that leveraging advanced architectures and attention mechanisms can substantially improve visual localization performance, even under challenging conditions. This research highlights the potential of integrating deep learning and graph neural networks in practical localization tasks

Read Full Paperexternally

Perguntar à IA

Bookmark

View Full Paper