Reference Pose Generation for Long-term Visual Localization via Learned Features and View Synthesis
This work addresses the need for high-quality reference poses in autonomous driving and augmented reality, offering a scalable solution to replace error-prone manual or SfM-based methods, though it is incremental as it builds on existing datasets and techniques.
The paper tackles the problem of generating accurate reference poses for visual localization under varying conditions by proposing a semi-automated method using learned features and view synthesis, resulting in up to 47% improvement in localization performance on the Aachen Day-Night dataset.
Visual Localization is one of the key enabling technologies for autonomous driving and augmented reality. High quality datasets with accurate 6 Degree-of-Freedom (DoF) reference poses are the foundation for benchmarking and improving existing methods. Traditionally, reference poses have been obtained via Structure-from-Motion (SfM). However, SfM itself relies on local features which are prone to fail when images were taken under different conditions, e.g., day/ night changes. At the same time, manually annotating feature correspondences is not scalable and potentially inaccurate. In this work, we propose a semi-automated approach to generate reference poses based on feature matching between renderings of a 3D model and real images via learned features. Given an initial pose estimate, our approach iteratively refines the pose based on feature matches against a rendering of the model from the current pose estimate. We significantly improve the nighttime reference poses of the popular Aachen Day-Night dataset, showing that state-of-the-art visual localization methods perform better (up to $47\%$) than predicted by the original reference poses. We extend the dataset with new nighttime test images, provide uncertainty estimates for our new reference poses, and introduce a new evaluation criterion. We will make our reference poses and our framework publicly available upon publication.