Semantic Pose Verification for Outdoor Visual Localization with Self-supervised Contrastive Learning

TL;DR

This paper introduces a self-supervised contrastive learning approach to estimate semantic similarity for outdoor visual localization, improving pose verification and increasing top-1 recall by 2% under challenging appearance changes.

Contribution

It presents a novel self-supervised contrastive learning method for semantic similarity estimation to enhance outdoor visual localization accuracy.

Findings

Semantic similarity estimation outperforms pixel-level methods.

Pose verification with semantic similarity improves top-1 recall to 0.90.

Contrastive learning-based approach is robust to appearance changes.

Abstract

Any city-scale visual localization system has to overcome long-term appearance changes, such as varying illumination conditions or seasonal changes between query and database images. Since semantic content is more robust to such changes, we exploit semantic information to improve visual localization. In our scenario, the database consists of gnomonic views generated from panoramic images (e.g. Google Street View) and query images are collected with a standard field-of-view camera at a different time. To improve localization, we check the semantic similarity between query and database images, which is not trivial since the position and viewpoint of the cameras do not exactly match. To learn similarity, we propose training a CNN in a self-supervised fashion with contrastive learning on a dataset of semantically segmented images. With experiments we showed that this semantic similarity estimation approach works better than measuring the similarity at pixel-level. Finally, we used the semantic similarity scores to verify the retrievals obtained by a state-of-the-art visual localization method and observed that contrastive learning-based pose verification increases top-1 recall value to 0.90 which corresponds to a 2% improvement.