A Real-Time Fusion Framework for Long-term Visual Localization
This work addresses visual localization for robotics applications, but it is incremental as it builds on existing fusion methods with minor improvements.
The paper tackles the problem of visual localization under challenging conditions like motion blur and illumination changes by proposing a client-server architecture that fuses global and local pose estimations, achieving competitive performance on benchmarks such as 4Seasons and OpenLORIS.
Visual localization is a fundamental task that regresses the 6 Degree Of Freedom (6DoF) poses with image features in order to serve the high precision localization requests in many robotics applications. Degenerate conditions like motion blur, illumination changes and environment variations place great challenges in this task. Fusion with additional information, such as sequential information and Inertial Measurement Unit (IMU) inputs, would greatly assist such problems. In this paper, we present an efficient client-server visual localization architecture that fuses global and local pose estimations to realize promising precision and efficiency. We include additional geometry hints in mapping and global pose regressing modules to improve the measurement quality. A loosely coupled fusion policy is adopted to leverage the computation complexity and accuracy. We conduct the evaluations on two typical open-source benchmarks, 4Seasons and OpenLORIS. Quantitative results prove that our framework has competitive performance with respect to other state-of-the-art visual localization solutions.