Learning to Detect Mobile Objects from LiDAR Scans Without Labels
Learning to Detect Mobile Objects from LiDAR Scans Without Labels
复制标题
DOI：
10.1109/cvpr52688.2022.00120
复制
发表时间：
2022-03
期刊：
影响因子：
--
通讯作者：
Yurong You;Katie Luo;Cheng Perng Phoo;Wei-Lun Chao;Wen Sun;Bharath Hariharan;M. Campbell;Kilian Q. Weinberger
中科院分区：
文献类型：
--
作者：
Yurong You;Katie Luo;Cheng Perng Phoo;Wei-Lun Chao;Wen Sun;Bharath Hariharan;M. Campbell;Kilian Q. Weinberger
目前用于自动驾驶的3D物体探测器几乎完全是在人类注释的数据上训练的。虽然这些数据的质量很高，但是生成这些数据很费力，而且成本很高，将它们限制在几个特定的位置和对象类型中。本文提出了一种完全基于未标记数据的替代方法，这种数据可以在地球上几乎任何地方廉价而丰富地收集。我们的方法利用几个简单的常识性启发式来创建一组初始的近似种子标签。例如，相关的交通参与者通常不会在同一路线的多次遍历中持久存在，不会飞行，也不会在地下。我们证明了这些种子标签是非常有效的，通过重复的自我训练，在没有单个人类注释标签的情况下，引导出一个惊人的精确检测器。代码可从https://github.com/YurongYou/MODEST获得。
Current 3D object detectors for autonomous driving are almost entirely trained on human-annotated data. Although of high quality, the generation of such data is laborious and costly, restricting them to a few specific locations and object types. This paper proposes an alternative approach entirely based on unlabeled data, which can be collected cheaply and in abundance almost everywhere on earth. Our approach leverages several simple common sense heuristics to create an initial set of approximate seed labels. For example, relevant traffic participants are generally not persistent across multiple traversals of the same route, do not fly, and are never under ground. We demonstrate that these seed labels are highly effective to bootstrap a surprisingly accurate detector through repeated self-training without a single human annotated label. Code is available at https://github.com/YurongYou/MODEST.