Detailed Information

Cited 3 time in webofscience Cited 0 time in scopus
Metadata Downloads

Encoding Pose Features to Images With Data Augmentation for 3-D Action Recognition

Authors
Huynh-The, ThienHua, Cam-HaoKim, Dong-Seong
Issue Date
May-2020
Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Keywords
Data augmentation; deep convolutional neural networks (DCNNs); human action recognition; pose feature to image (PoF2I) encoding technique
Citation
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, v.16, no.5, pp.3100 - 3111
Journal Title
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
Volume
16
Number
5
Start Page
3100
End Page
3111
URI
https://scholarworks.bwise.kr/kumoh/handle/2020.sw.kumoh/17607
DOI
10.1109/TII.2019.2910876
ISSN
1551-3203
Abstract
Recently, numerous methods have been introduced for three-dimensional (3-D) action recognition using handcrafted feature descriptors coupled traditional classifiers. However, they cannot learn high-level features of a whole skeleton sequence exhaustively. In this paper, a novel encoding technique-namely, pose feature to image (PoF2I), is introduced to transform the pose features of joint-joint distance and orientation to color pixels. By concatenating the features of all skeleton frames in a sequence, a color image is generated to depict spatial joint correlations and temporal pose dynamics of an action appearance. The strategy of end-to-end fine-tuning a pretrained deep convolutional neural network, which completely capture multiple high-level features at multiscale action representation, is implemented for learning recognition models. We further propose an efficient data augmentation mechanism for informative enrichment and overfitting prevention. The experimental results on six challenging 3-D action recognition datasets demonstrate that the proposed method outperforms state-of-the-art approaches.
Files in This Item:
There are no files associated with this item.
Appears in
Collections:
ETC > 1. Journal Articles

qrcode

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

Related Researcher

Researcher

KIM, DONG SEONG
전자공학부
Read more

Views & Downloads

MENU