USIP: Unsupervised Stable Interest Point Detection from 3D Point Clouds
This addresses the need for reliable keypoint detection in 3D computer vision applications like robotics and autonomous driving, offering an unsupervised approach that outperforms prior work.
The paper tackles the problem of detecting stable interest points from 3D point clouds without ground truth data, achieving significantly higher repeatability than existing methods in experiments on simulated and real-world datasets.
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 of our USIP detector 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