平行点云:虚实互动的点云生成与三维模型进化方法  

Parallel Point Clouds:Point Clouds Generation and 3D Model Evolution via Virtual-real Interaction

作　　者：田永林 沈宇 李强 王飞跃[2,4] TIAN Yong-Lin;SHEN Yu;LI Qiang;WANG Fei-Yue(Department of Automation,University of Science and Technology of China,Hefei 230027;State Key Laboratory of Management and Control for Complex Systems,Institute of Automation,Chinese Academy of Sciences,Beijing 100190;School of Artiflcial Intelligence,University of Chinese Academy of Sciences,Beijing 100049;Qingdao Academy of Intelligent Industries,Qingdao 266000)

机构地区：[1]中国科学技术大学自动化系,合肥230027 [2]中国科学院自动化研究所复杂系统管理与控制国家重点实验室,北京100190 [3]中国科学院大学人工智能学院,北京100049 [4]青岛智能产业技术研究院,青岛266000

出　　处：《自动化学报》2020年第12期2572-2582,共11页Acta Automatica Sinica

基　　金：国家自然科学基金重点项目(61533019);英特尔智能网联汽车大学合作研究中心项目(ICRI-IACV);国家自然科学基金项目联合基金(U1811463);广州市智能网联汽车重大科技专项(202007050002)资助。

摘　　要：三维信息的提取在自动驾驶等智能交通场景中正发挥着越来越重要的作用,为了解决以激光雷达为主的深度传感器在数据采集方面面临的成本高、样本覆盖不全面等问题,本文提出了平行点云的框架.利用人工定义场景获取虚拟点云数据,通过计算实验训练三维模型,借助平行执行对模型性能进行测试,并将结果反馈至数据生成和模型训练过程.通过不断地迭代,使三维模型得到充分评估并不断进化.在平行点云的框架下,我们以三维目标检测为例,通过闭环迭代,构建了虚实结合的点云数据集,在无需人工标注的情况下,可达到标注数据训练模型精度的72%.The extraction of 3D information is playing an increasingly important role in intelligent traffic scenes such as autonomous driving.In order to solve the problems faced by LiDAR sensor such as the high cost and incomplete coverage of possible scenarios,this paper proposes parallel point clouds and its framework.For parallel point clouds,virtual point clouds are obtained by building artiflcial scenes.Then 3D models are trained through computational experiments and tested by parallel execution.The evaluation results are fed back to the data generation and the training process of 3D models.Through continuous iteration,3D models can be fully evaluated and updated.Under the framework of Parallel Point Clouds,we take the 3D object detection as an example and build a point clouds dataset in a closed-loop manner.Without human annotation,it can be used to effectively train the detection model which can achieve the 72%of the performance of model trained with annotated data.

关 键 词：平行点云 虚实互动 三维视觉模型 三维目标检测 

分 类 号：U463.6[机械工程—车辆工程] U495[交通运输工程—载运工具运用工程] TN958.98[交通运输工程—道路与铁道工程]