arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2110.04076cs.CVcs.RO
收藏

三维时空卷积网络自监督点云预测

Self-supervised Point Cloud Prediction Using 3D Spatio-temporal Convolutional Networks

University of Bonn(波恩大学)

Benedikt Mersch, Xieyuanli Chen, Jens Behley, Cyrill Stachniss

摘要:

利用过去的三维激光雷达扫描来预测未来的点云是一种很有前途的方法,自主移动系统,以实现前瞻性的状态估计,避免碰撞,和规划。在本文中,我们解决的问题,预测未来的三维激光雷达点云给定的序列过去的激光雷达扫描。在传感器级别上估计未来场景不需要像定位或跟踪系统中那样的任何先前步骤,并且可以进行自我监督训练。我们提出了一种端到端的方法,利用每个3D激光雷达扫描的2D范围图像表示,并连接范围图像序列以获得3D张量。基于这样的张量,我们开发了一个编码器-解码器架构,使用3D卷积来联合聚合场景的空间和时间信息,并预测未来的3D点云。我们在多个数据集上评估了我们的方法,实验结果表明,我们的方法优于现有的点云预测架构,并且可以很好地推广到新的、看不见的环境,而无需进行额外的微调。我们的方法在线运行,并且比10 Hz的常见LiDAR帧速率更快。

英文摘要:

Exploiting past 3D LiDAR scans to predict future point clouds is a promising method for autonomous mobile systems to realize foresighted state estimation, collision avoidance, and planning. In this paper, we address the problem of predicting future 3D LiDAR point clouds given a sequence of past LiDAR scans. Estimating the future scene on the sensor level does not require any preceding steps as in localization or tracking systems and can be trained self-supervised. We propose an end-to-end approach that exploits a 2D range image representation of each 3D LiDAR scan and concatenates a sequence of range images to obtain a 3D tensor. Based on such tensors, we develop an encoder-decoder architecture using 3D convolutions to jointly aggregate spatial and temporal information of the scene and to predict the future 3D point clouds. We evaluate our method on multiple datasets and the experimental results suggest that our method outperforms existing point cloud prediction architectures and generalizes well to new, unseen environments without additional fine-tuning. Our method operates online and is faster than the common LiDAR frame rate of 10 Hz.

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。

↑