EgoNN: Egocentric Neural Network for Point Cloud Based 6DoF Relocalization at the City Scale
This addresses the problem of precise localization for autonomous vehicles or robotics in urban environments, presenting an incremental improvement with a novel method for descriptor extraction.
The paper tackles 6DoF relocalization at city scale using point clouds from LiDAR, proposing a deep neural network to extract global and local descriptors for a two-stage approach that retrieves coarse position and estimates pose, achieving efficient processing of tens of thousands of points.
The paper presents a deep neural network-based method for global and local descriptors extraction from a point cloud acquired by a rotating 3D LiDAR. The descriptors can be used for two-stage 6DoF relocalization. First, a course position is retrieved by finding candidates with the closest global descriptor in the database of geo-tagged point clouds. Then, the 6DoF pose between a query point cloud and a database point cloud is estimated by matching local descriptors and using a robust estimator such as RANSAC. Our method has a simple, fully convolutional architecture based on a sparse voxelized representation. It can efficiently extract a global descriptor and a set of keypoints with local descriptors from large point clouds with tens of thousand points. Our code and pretrained models are publicly available on the project website.