Visual Localization Based on Multiple Maps

This paper proposes a multi-map based visual localization method for image sequences. Given multiple single-map based localization results, we combine them with SLAM to estimate robust and accurate camera poses under challenging conditions. Our method comprises three modules connected in a sequence. First, we reconstruct multiple reference maps using the Structure-from-Motion technique, one map for each reference sequence. A single-image-based localization pipeline is performed to estimate 6-DoF camera poses for each query image, one for each map. Second, a consensus set maximization module is proposed to select the best camera poses from multi-map poses, estimating one 6-DoF camera pose for each query image. Finally, a robust pose refinement module is proposed to optimize 6-DoF camera poses of query images, combining map-based localization and local SLAM information. Experiments show that the proposed pipeline achieves state-of-the-art performance on challenging map-based localization benchmarks. Demonstrating the broad applicability of our method, we obtained first place in the challenge of Map-Based Localization for Autonomous Driving at ECCV2022.

Authors

• Yukai Lin
• Liu Liu
• Xiao Liang
• Jiangwei Li

Keywords

• Location awareness
• Visualization
• Simultaneous localization and mapping
• Pipelines
• Pose estimation
• Lighting
• Cameras
• Multiple Mapping
• Visual Localization
• Reference Genome
• Local Information
• Local Method
• Visual Methods
• Autonomous Vehicles
• Challenging Conditions
• Reference Map
• Camera Pose
• Query Image
• Environmental Changes
• Objective Function
• Localization Accuracy
• Nonlinear Programming
• Query Sequence
• Consecutive Frames
• Feature Matching
• Relative Pose
• Bundle Adjustment
• Reprojection Error
• Pose Error
• Global Localization
• Feature Tracking
• Merging Method
• Global Error
• Robust Localization
• Scene Changes