Recurrent Multi-view Alignment Network for Unsupervised Surface Registration

TL;DR

This paper introduces a novel recurrent multi-view alignment network that enables unsupervised non-rigid surface registration by representing transformations as combinations of rigid motions and using a differentiable multi-view shape similarity loss.

Contribution

It proposes a new non-rigid registration method using a point-wise rigid transformation combination and a differentiable multi-view shape loss for end-to-end training without labels.

Findings

Outperforms previous state-of-the-art methods significantly.

Effectively handles high degrees of freedom in non-rigid registration.

Demonstrates robustness across multiple datasets.

Abstract

Learning non-rigid registration in an end-to-end manner is challenging due to the inherent high degrees of freedom and the lack of labeled training data. In this paper, we resolve these two challenges simultaneously. First, we propose to represent the non-rigid transformation with a point-wise combination of several rigid transformations. This representation not only makes the solution space well-constrained but also enables our method to be solved iteratively with a recurrent framework, which greatly reduces the difficulty of learning. Second, we introduce a differentiable loss function that measures the 3D shape similarity on the projected multi-view 2D depth images so that our full framework can be trained end-to-end without ground truth supervision. Extensive experiments on several different datasets demonstrate that our proposed method outperforms the previous state-of-the-art by a large margin. The source codes are available at https://github.com/WanquanF/RMA-Net.