HumanReg: Self-supervised Non-rigid Registration of Human Point Cloud
This addresses the problem of expensive annotation requirements in human point cloud registration for computer vision and robotics applications, offering an incremental improvement with self-supervised training.
The paper tackles non-rigid registration of human point clouds by introducing HumanReg, a self-supervised framework that uses body priors and novel loss functions, achieving state-of-the-art performance on the CAPE-512 dataset and demonstrating effectiveness on a more challenging real-world dataset.
In this paper, we present a novel registration framework, HumanReg, that learns a non-rigid transformation between two human point clouds end-to-end. We introduce body prior into the registration process to efficiently handle this type of point cloud. Unlike most exsisting supervised registration techniques that require expensive point-wise flow annotations, HumanReg can be trained in a self-supervised manner benefiting from a set of novel loss functions. To make our model better converge on real-world data, we also propose a pretraining strategy, and a synthetic dataset (HumanSyn4D) consists of dynamic, sparse human point clouds and their auto-generated ground truth annotations. Our experiments shows that HumanReg achieves state-of-the-art performance on CAPE-512 dataset and gains a qualitative result on another more challenging real-world dataset. Furthermore, our ablation studies demonstrate the effectiveness of our synthetic dataset and novel loss functions. Our code and synthetic dataset is available at https://github.com/chenyifanthu/HumanReg.