From CAD models to soft point cloud labels: An automatic annotation pipeline for cheaply supervised 3D semantic segmentation

TL;DR

This paper introduces an automatic annotation pipeline that leverages CAD models to generate accurate, cost-effective point-wise labels for 3D point cloud segmentation, reducing manual effort and improving training data quality.

Contribution

The authors present a fully automatic labeling method that produces soft point-wise labels from CAD models, enhancing 3D segmentation training without manual annotation.

Findings

Automatic labels are accurate and reduce annotation time.

Soft labels improve segmentation performance over hard labels.

Method is effective on industrial and indoor scene datasets.

Abstract

We propose a fully automatic annotation scheme that takes a raw 3D point cloud with a set of fitted CAD models as input and outputs convincing point-wise labels that can be used as cheap training data for point cloud segmentation. Compared with manual annotations, we show that our automatic labels are accurate while drastically reducing the annotation time and eliminating the need for manual intervention or dataset-specific parameters. Our labeling pipeline outputs semantic classes and soft point-wise object scores, which can either be binarized into standard one-hot-encoded labels, thresholded into weak labels with ambiguous points left unlabeled, or used directly as soft labels during training. We evaluate the label quality and segmentation performance of PointNet++ on a dataset of real industrial point clouds and Scan2CAD, a public dataset of indoor scenes. Our results indicate that reducing supervision in areas that are more difficult to label automatically is beneficial compared with the conventional approach of naively assigning a hard "best guess" label to every point.