Labelling point clouds fully is highly time-consuming and costly. As larger point cloud datasets with billions of points become more common, we ask whether the full annotation is even necessary, demonstrating that existing baselines designed under a fully annotated assumption only degrade slightly even when faced with 1% random point annotations. However, beyond this point, e.g., at 0.1% annotations, segmentation accuracy is unacceptably low. We observe that, as point clouds are samples of the 3D world, the distribution of points in a local neighborhood is relatively homogeneous, exhibiting strong semantic similarity. Motivated by this, we propose a new weak supervision method to implicitly augment highly sparse supervision signals. Extensive experiments demonstrate the proposed Semantic Query Network (SQN) achieves promising performance on seven large-scale open datasets under weak supervision schemes, while requiring only 0.1% randomly annotated points for training, greatly reducing annotation cost and effort. The code is available at https://github.com/QingyongHu/SQN.
@article{arxiv.2104.04891,
title = {SQN: Weakly-Supervised Semantic Segmentation of Large-Scale 3D Point Clouds},
author = {Qingyong Hu and Bo Yang and Guangchi Fang and Yulan Guo and Ales Leonardis and Niki Trigoni and Andrew Markham},
journal= {arXiv preprint arXiv:2104.04891},
year = {2023}
}