SuperGF: Unifying Local and Global Features for Visual Localization

TL;DR

SuperGF is a transformer-based model that unifies local and global features for visual localization, improving accuracy and efficiency over previous methods by operating directly on image-matching features.

Contribution

It introduces a novel unified approach for local and global feature extraction in visual localization, enhancing accuracy and computational efficiency.

Findings

SuperGF outperforms existing methods in accuracy and efficiency.

It works effectively with various local feature types.

Experimental results validate its advantages in real-world scenarios.

Abstract

Advanced visual localization techniques encompass image retrieval challenges and 6 Degree-of-Freedom (DoF) camera pose estimation, such as hierarchical localization. Thus, they must extract global and local features from input images. Previous methods have achieved this through resource-intensive or accuracy-reducing means, such as combinatorial pipelines or multi-task distillation. In this study, we present a novel method called SuperGF, which effectively unifies local and global features for visual localization, leading to a higher trade-off between localization accuracy and computational efficiency. Specifically, SuperGF is a transformer-based aggregation model that operates directly on image-matching-specific local features and generates global features for retrieval. We conduct experimental evaluations of our method in terms of both accuracy and efficiency, demonstrating its advantages over other methods. We also provide implementations of SuperGF using various types of local features, including dense and sparse learning-based or hand-crafted descriptors.