Unsupervised Point Cloud Registration via Salient Points Analysis (SPA)
This addresses the need for efficient and unsupervised registration in computer vision, offering advantages over supervised deep learning methods in terms of training time and model size, though it is incremental as it builds on existing techniques like PointHop++.
The paper tackles the problem of point cloud registration by proposing an unsupervised method called salient points analysis (SPA), which uses a small subset of salient points to achieve effective registration, as demonstrated on the ModelNet-40 dataset with seen and unseen classes and noisy data.
An unsupervised point cloud registration method, called salient points analysis (SPA), is proposed in this work. The proposed SPA method can register two point clouds effectively using only a small subset of salient points. It first applies the PointHop++ method to point clouds, finds corresponding salient points in two point clouds based on the local surface characteristics of points and performs registration by matching the corresponding salient points. The SPA method offers several advantages over the recent deep learning based solutions for registration. Deep learning methods such as PointNetLK and DCP train end-to-end networks and rely on full supervision (namely, ground truth transformation matrix and class label). In contrast, the SPA is completely unsupervised. Furthermore, SPA's training time and model size are much less. The effectiveness of the SPA method is demonstrated by experiments on seen and unseen classes and noisy point clouds from the ModelNet-40 dataset.