USIP: Unsupervised Stable Interest Point Detection From 3D Point Clouds

• Published in: 2019 IEEE/CVF International Conference on Computer Vision (ICCV) pp. 361 - 370
• Main Authors: Li, Jiaxin, Lee, Gim Hee
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.10.2019
• Subjects: Detectors; Probabilistic logic; Proposals; Solid modeling; Three-dimensional displays; Training; Uncertainty
• ISSN: 2380-7504
• DOI: 10.1109/ICCV.2019.00045

In this paper, we propose the USIP detector: an Unsupervised Stable Interest Point detector that can detect highly repeatable and accurately localized keypoints from 3D point clouds under arbitrary transformations without the need for any ground truth training data. Our USIP detector consists of a feature proposal network that learns stable keypoints from input 3D point clouds and their respective transformed pairs from randomly generated transformations. We provide degeneracy analysis and suggest solutions to prevent it. We encourage high repeatability and accurate localization of the keypoints with a probabilistic chamfer loss that minimizes the distances between the detected keypoints from the training point cloud pairs. Extensive experimental results of repeatability tests on several simulated and real-world 3D point cloud datasets from Lidar, RGB-D and CAD models show that our USIP detector significantly outperforms existing hand-crafted and deep learning-based 3D keypoint detectors. Our code is available at the project website. https://github.com/lijx10/USIP.