Self-Supervised Point Set Local Descriptors for Point Cloud Registration

Descriptors play an important role in point cloud registration. The current state-of-the-art resorts to the high regression capability of deep learning. However, recent deep learning-based descriptors require different levels of annotation and selection of patches, which make the model hard to migrate to new scenarios. In this work, we learn local registration descriptors for point clouds in a self-supervised manner. In each iteration of the training, the input of the network is merely one unlabeled point cloud. Thus, the whole training requires no manual annotation and manual selection of patches. In addition, we propose to involve keypoint sampling into the pipeline, which further improves the performance of our model. Our experiments demonstrate the capability of our self-supervised local descriptor to achieve even better performance than the supervised model, while being easier to train and requiring no data labeling.

Automated summary

In this work, self-supervised learning of local registration descriptors for point clouds is achieved, eliminating the need for manual annotation and patch selection. Key point sampling is also incorporated into the pipeline, further enhancing model performance. Our experiments show that the self-supervised local descriptor can outperform the supervised model while being easier to train and requiring no data labeling.