CASIA OpenIR  > 智能感知与计算研究中心
Beyond Joints: Learning Representations From Primitive Geometries for Skeleton-Based Action Recognition and Detection
Wang Hongsong(王洪松)1,2,3,4; Wang Liang(王亮)1,2,3,4
Source PublicationIEEE Transactions on Image Processing
2018
Volume27Issue:9Pages:4382 - 4394
Abstract

Recently, skeleton-based action recognition becomes popular owing to the development of cost-effective depth sensors and fast pose estimation algorithms. Traditional methods based on pose descriptors often fail on large-scale datasets due to the limited representation of engineered features. Recent recurrent neural networks (RNN) based approaches mostly focus on the temporal evolution of body joints and neglect the geometric relations. In this paper, we aim to leverage the geometric relations among joints for action recognition. We introduce three primitive geometries: joints, edges and surfaces. Accordingly, a generic end-to-end RNN based network is designed to accommodate the three inputs. For action recognition, a novel viewpoint transformation layer and temporal dropout layers are utilized in the RNN based network to learn robust representations. And for action detection, we first perform frame-wise action classification, then exploit a novel multi-scale sliding window algorithm. Experiments on the large-scale 3D action recognition benchmark datasets show that joints, edges and surfaces are effective and complementary for different actions. Our approaches dramatically outperform the existing state-of-the-art methods for both tasks of action recognition and action detection.

KeywordSkeleton-based Action Recognition Geometric Relations Viewpoint Transformation Action Detection
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/21061
Collection智能感知与计算研究中心
Affiliation1.Center for Research on Intelligent Perception and Computing (CRIPAC)
2.National Laboratory of Pattern Recognition (NLPR)
3.Institute of Automation, Chinese Academy of Sciences
4.University of Chinese Academy of Sciences
Recommended Citation
GB/T 7714
Wang Hongsong,Wang Liang. Beyond Joints: Learning Representations From Primitive Geometries for Skeleton-Based Action Recognition and Detection[J]. IEEE Transactions on Image Processing,2018,27(9):4382 - 4394.
APA Wang Hongsong,&Wang Liang.(2018).Beyond Joints: Learning Representations From Primitive Geometries for Skeleton-Based Action Recognition and Detection.IEEE Transactions on Image Processing,27(9),4382 - 4394.
MLA Wang Hongsong,et al."Beyond Joints: Learning Representations From Primitive Geometries for Skeleton-Based Action Recognition and Detection".IEEE Transactions on Image Processing 27.9(2018):4382 - 4394.
Files in This Item: Download All
File Name/Size DocType Version Access License
Beyond Joints Learni(2429KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Wang Hongsong(王洪松)]'s Articles
[Wang Liang(王亮)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Wang Hongsong(王洪松)]'s Articles
[Wang Liang(王亮)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Wang Hongsong(王洪松)]'s Articles
[Wang Liang(王亮)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Beyond Joints Learning Representations From Primitive Geometries for Skeleton-Based Action Recognition and Detection.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.