arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1807.06288cs.CV
收藏

PointSeg:基于三维LiDAR点云的实时的语义分割

PointSeg: Real-Time Semantic Segmentation Based on 3D LiDAR Point Cloud

The Hong Kong University of Science and Technology(香港科技大学);Nanjing University of Aeronautics and Astronautics(南京航空航天大学)

Yuan Wang, Tianyue Shi, Peng Yun, Lei Tai, Ming Liu

摘要:

在本文中,我们提出了PointSeg,一个实时的【形】的端到端的语义分割方法的基础上球形图像的道路对象。我们将从3D LiDAR点云转换的球形图像作为卷积神经网络(CNN)的输入,以预测逐点语义地图。为了使PointSeg适用于移动系统,我们基于轻量级网络SqueezeNet构建了模型,并进行了一些改进。它在内存开销和预测性能之间保持了良好的平衡。我们的模型是在从KITTI 3D目标检测数据集投影的球形图像和标签掩模上训练的。实验表明,PointSeg在单GPU 1080ti上可以达到90fps的竞争精度。这使得它与自动驾驶应用非常兼容。

英文摘要:

In this paper, we propose PointSeg, a real-time end-to-end semantic segmentation method for road-objects based on spherical images. We take the spherical image, which is transformed from the 3D LiDAR point clouds, as input of the convolutional neural networks (CNNs) to predict the point-wise semantic map. To make PointSeg applicable on a mobile system, we build the model based on the light-weight network, SqueezeNet, with several improvements. It maintains a good balance between memory cost and prediction performance. Our model is trained on spherical images and label masks projected from the KITTI 3D object detection dataset. Experiments show that PointSeg can achieve competitive accuracy with 90fps on a single GPU 1080ti. which makes it quite compatible for autonomous driving applications.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑