6-DoF Pose Refinement via Sparse-to-Dense Feature-Metric Optimization

Abstract

In this report, we detail our submission to the CVPR 2020 visual localization challenge. Previous work on sparse-to-dense matching showed outstanding robustness to extreme conditions, but lacks precision due to the low resolution of deep feature maps. We introduce a simple algorithm to refine the estimated pose based on the feature-metric error, and demonstrate improved localization accuracy. This, combined with better feature selection, results in state-of-art night localization on the RobotCar dataset.

Publication
CVPR 2020 Workshop on Long-Term Visual Localization, Visual Odometry and Geometric and Learning-based SLAM
Ajay Unagar
Ajay Unagar
Applied Research Scientist

My research interests include 3D Vision, Reinforcement Learning, and Robotics.

Related