Velocity-to-velocity human motion forecasting

doi:10.1016/j.patcog.2021.108424

CASIA OpenIR > 智能感知与计算研究中心

	Velocity-to-velocity human motion forecasting
	Wang, Hongsong1 ; Wang, Liang2 ; Feng, Jiashi 1; Zhou, Daquan 1
发表期刊	PATTERN RECOGNITION
ISSN	0031-3203
	2022-04-01
卷号	124 页码:11
通讯作者	Wang, Hongsong(hongsongsui@gmail.com)
摘要	Forecasting human motion from a sequence of human poses is an important problem in the fields of computer vision and robotics. Most previous approaches merely consider learning the temporal dynamics of body joints or joint angles, while neglect derivatives of body joints (i.e., pose velocities) which could reasonably reduce noise impact and improve stability. To exploit the benefits of pose velocities, we propose the velocity-to-velocity learning paradigm for human motion prediction which attempts to directly build the sequence-to-sequence model in the velocity space. Two variant architectures based on recurrent encoder-decoder networks are introduced under this paradigm. Considering human motion as kinematics of rigid bodies, joint angles which denote transformation are the computations of inverse kinematics. Accordingly, a novel loss function in terms of rotation matrices is designed during training for human motion prediction through a rotation matrix transformation (RMT) layer. Finally, we present an effective training algorithm which exploits sequence transformation to improve model generalization. Our approaches substantially outperform state-of-the-art approaches on two large-scale datasets, Human3.6M and CMU Motion Capture, for both short-term prediction and long-term prediction. In particular, our model can competently forecast human-like and meaningful poses up to 10 0 0 milliseconds. The code is available on GitHub: https://github.com/hongsong-wang/RNN _ based _ human _ motion _ prediction .(c) 2021 Elsevier Ltd. All rights reserved.
关键词	Human motion prediction Action anticipation
DOI	10.1016/j.patcog.2021.108424
关键词[WOS]	ACTION RECOGNITION
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[61633021] ; National Natural Science Foundation of China[61721004] ; National Natural Science Foundation of China[U1803261] ; Shandong Provincial Key Research and Development Program[2019JZZY010119] ; CAS-AIR
项目资助者	National Natural Science Foundation of China ; Shandong Provincial Key Research and Development Program ; CAS-AIR
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:000776697500005
出版者	ELSEVIER SCI LTD
引用统计	被引频次：2[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/48314
专题	智能感知与计算研究中心
通讯作者	Wang, Hongsong
作者单位	1.Natl Univ Singapore, Singapore, Singapore 2.Chinese Acad Sci, Univ Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Wang, Hongsong,Wang, Liang,Feng, Jiashi,et al. Velocity-to-velocity human motion forecasting[J]. PATTERN RECOGNITION,2022,124:11.
APA	Wang, Hongsong,Wang, Liang,Feng, Jiashi,&Zhou, Daquan.(2022).Velocity-to-velocity human motion forecasting.PATTERN RECOGNITION,124,11.
MLA	Wang, Hongsong,et al."Velocity-to-velocity human motion forecasting".PATTERN RECOGNITION 124(2022):11.