VKFPos: A Learning-Based Monocular Positioning with Variational Bayesian Extended Kalman Filter Integration
This work addresses positioning challenges in computer vision, offering an incremental improvement for applications like robotics or autonomous navigation.
The paper tackles monocular positioning by proposing VKFPos, which integrates Absolute and Relative Pose Regression with an Extended Kalman Filter using variational Bayesian inference, achieving superior accuracy in temporal positioning compared to existing methods.
This paper addresses the challenges in learning-based monocular positioning by proposing VKFPos, a novel approach that integrates Absolute Pose Regression (APR) and Relative Pose Regression (RPR) via an Extended Kalman Filter (EKF) within a variational Bayesian inference framework. Our method shows that the essential posterior probability of the monocular positioning problem can be decomposed into APR and RPR components. This decomposition is embedded in the deep learning model by predicting covariances in both APR and RPR branches, allowing them to account for associated uncertainties. These covariances enhance the loss functions and facilitate EKF integration. Experimental evaluations on both indoor and outdoor datasets show that the single-shot APR branch achieves accuracy on par with state-of-the-art methods. Furthermore, for temporal positioning, where consecutive images allow for RPR and EKF integration, VKFPos outperforms temporal APR and model-based integration methods, achieving superior accuracy.