Deep Weighted Consensus: Dense correspondence confidence maps for 3D shape registration
This work solves the problem of robust 3D shape registration for real-world applications involving large rotations and outliers, which is a significant improvement over current models.
This paper addresses the challenge of rigid alignment between point clouds, particularly under large rotations and high noise levels. The authors propose a new paradigm based on learnable weighted consensus, which achieves a fundamental boost in performance compared to existing methods.
We present a new paradigm for rigid alignment between point clouds based on learnable weighted consensus which is robust to noise as well as the full spectrum of the rotation group. Current models, learnable or axiomatic, work well for constrained orientations and limited noise levels, usually by an end-to-end learner or an iterative scheme. However, real-world tasks require us to deal with large rotations as well as outliers and all known models fail to deliver. Here we present a different direction. We claim that we can align point clouds out of sampled matched points according to confidence level derived from a dense, soft alignment map. The pipeline is differentiable, and converges under large rotations in the full spectrum of SO(3), even with high noise levels. We compared the network to recently presented methods such as DCP, PointNetLK, RPM-Net, PRnet, and axiomatic methods such as ICP and Go-ICP. We report here a fundamental boost in performance.