HR-APR: APR-agnostic Framework with Uncertainty Estimation and Hierarchical Refinement for Camera Relocalisation
This work addresses the issue of unreliable pose predictions in camera relocalization for robotics and AR/VR applications, offering a flexible and efficient solution that is incremental over existing uncertainty-aware methods.
The paper tackles the problem of unstable accuracy in Absolute Pose Regressors (APRs) for camera relocalization by introducing HR-APR, an APR-agnostic framework that estimates uncertainty via cosine similarity and uses it for hierarchical refinement, reducing computational overhead by 27.4% and 15.2% on two datasets while maintaining state-of-the-art accuracy.
Absolute Pose Regressors (APRs) directly estimate camera poses from monocular images, but their accuracy is unstable for different queries. Uncertainty-aware APRs provide uncertainty information on the estimated pose, alleviating the impact of these unreliable predictions. However, existing uncertainty modelling techniques are often coupled with a specific APR architecture, resulting in suboptimal performance compared to state-of-the-art (SOTA) APR methods. This work introduces a novel APR-agnostic framework, HR-APR, that formulates uncertainty estimation as cosine similarity estimation between the query and database features. It does not rely on or affect APR network architecture, which is flexible and computationally efficient. In addition, we take advantage of the uncertainty for pose refinement to enhance the performance of APR. The extensive experiments demonstrate the effectiveness of our framework, reducing 27.4\% and 15.2\% of computational overhead on the 7Scenes and Cambridge Landmarks datasets while maintaining the SOTA accuracy in single-image APRs.