While multi-class 3D detectors are needed in many robotics applications, training them with fully labeled datasets can be expensive in labeling cost. An alternative approach is to have targeted single-class labels on disjoint data samples. In this paper, we are interested in training a multi-class 3D object detection model, while using these single-class labeled data. We begin by detailing the unique stance of our "Single-Class Supervision" (SCS) setting with respect to related concepts such as partial supervision and semi supervision. Then, based on the case study of training the multi-class version of Range Sparse Net (RSN), we adapt a spectrum of algorithms -- from supervised learning to pseudo-labeling -- to fully exploit the properties of our SCS setting, and perform extensive ablation studies to identify the most effective algorithm and practice. Empirical experiments on the Waymo Open Dataset show that proper training under SCS can approach or match full supervision training while saving labeling costs.

翻译：虽然在许多机器人应用中需要多级三维探测器,但用贴满标签的数据集培训它们的费用在标签成本方面可能非常昂贵。 另一种办法是在脱节数据样本中设定单一级标签。 在本文中,我们有兴趣培训多级三维物体探测模型,同时使用这些单级标签数据。 我们首先详细说明我们的“单级监督”在部分监督和半监督等相关概念方面的独特立场。 然后,根据培训多级版本的“区域松散网(RSN)”的案例研究,我们调整一系列算法 -- -- 从监督学习到假标签 -- -- 以充分利用我们的SCS设置特性,并进行广泛的调整研究以确定最有效的算法和做法。Waymo Open Dataset(Scustainalalalalal Acustoration)的实验表明,在SCSCS下进行的适当培训可以接近或匹配全面监督培训,同时节省标签成本。