Multi-Class 3D Object Detection with Single-Class Supervision

• Main Authors: Ye, Mao, Liu, Chenxi, Yao, Maoqing, Wang, Weiyue, Leng, Zhaoqi, Qi, Charles R, Anguelov, Dragomir
• Format: Journal Article
• Language: English
• Published: 11-05-2022
• Subjects: Computer Science - Computer Vision and Pattern Recognition; Computer Science - Robotics
• DOI: 10.48550/arxiv.2205.05703

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.