HumanReg: Self-supervised Non-rigid Registration of Human Point Cloud

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Yifan, Pan, Zhiyu, Zhong, Zhicheng, Guo, Wenxuan, Feng, Jianjiang, Zhou, Jie
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909094697238528
author Chen, Yifan
Pan, Zhiyu
Zhong, Zhicheng
Guo, Wenxuan
Feng, Jianjiang
Zhou, Jie
author_facet Chen, Yifan
Pan, Zhiyu
Zhong, Zhicheng
Guo, Wenxuan
Feng, Jianjiang
Zhou, Jie
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2312_05462
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle HumanReg: Self-supervised Non-rigid Registration of Human Point Cloud
Chen, Yifan
Pan, Zhiyu
Zhong, Zhicheng
Guo, Wenxuan
Feng, Jianjiang
Zhou, Jie
Computer Vision and Pattern Recognition
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.
title HumanReg: Self-supervised Non-rigid Registration of Human Point Cloud
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2312.05462