LO-Net: Deep Real-time Lidar Odometry

Abstract

We present a novel deep convolutional network pipeline, LO-Net, for real-time lidar odometry estimation. Unlike most existing lidar odometry (LO) estimations that go through individually designed feature selection, feature matching, and pose estimation pipeline, LO-Net can be trained in an end-to-end manner. With a new mask-weighted geometric constraint loss, LO-Net can effectively learn feature representation for LO estimation, and can implicitly exploit the sequential dependencies and dynamics in the data. We also design a scan-to-map module, which uses the geometric and semantic information learned in LO-Net, to improve the estimation accuracy. Experiments on benchmark datasets demonstrate that LO-Net outperforms existing learning based approaches and has similar accuracy with the state-of-the-art geometry-based approach, LOAM.

References in corpus (8)

• Adam: A Method for Stochastic Optimization
• SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
• DeepVO: Towards End-to-End Visual Odometry with Deep Recurrent Convolutional Neural Networks
• Unsupervised Learning of Depth and Ego-Motion from Video
• PointCNN: Convolution On $\mathcal{X}$-Transformed Points
• Unsupervised Learning of Geometry with Edge-aware Depth-Normal Consistency
• PointSeg: Real-Time Semantic Segmentation Based on 3D LiDAR Point Cloud
• Geometry-Aware Learning of Maps for Camera Localization