Structured RNN for human interaction

Understanding human activities has been an important research area in computer vision. Generally, the authors can model the human interactions as a temporal sequence with the transition in relationships of humans and objects. Besides, many studies have proved the effectiveness of long short-term mem...

Full description

Saved in:
Bibliographic Details
Published inIET computer vision Vol. 12; no. 6; pp. 817 - 825
Main Authors Truong, Anh Minh, Yoshitaka, Atsuo
Format Journal Article
LanguageEnglish
Published The Institution of Engineering and Technology 01.09.2018
Wiley
Subjects
CAD-120
computer vision
human activities
human interaction
long short-term memory
long-term temporal dependency problems
perceptrons
recurrent neural nets
Research Article
S-RNN
SBU-Kinect-Interaction
single-layer perceptron
spatio-temporal relationships
structured recurrent neural network
temporal sequence
perceptrons
human activities
recurrent neural nets
SBU-Kinect-Interaction
long-term temporal dependency problems
single-layer perceptron
human interaction
spatio-temporal relationships
computer vision
CAD-120
temporal sequence
structured recurrent neural network
S-RNN
long short-term memory
Online AccessGet full text

Cover

More Information
Summary:Understanding human activities has been an important research area in computer vision. Generally, the authors can model the human interactions as a temporal sequence with the transition in relationships of humans and objects. Besides, many studies have proved the effectiveness of long short-term memory (LSTM) on long-term temporal dependency problems. Here, the authors proposed a novel structured recurrent neural network (S-RNN) to model spatio-temporal relationships between human subjects and objects in daily human interactions. The authors represent the evolution of different components and the relationships between them over time by several subnets. Then, the hidden representations of those relations are fused and fed into the later layers to obtain the final hidden representation. The final prediction is carried out by the single-layer perceptron. The experimental results of different tasks on the CAD-120, SBU-Kinect-Interaction, multi-modal and multi-view and interactive, and NTU RGB+D data sets showed advantages of the proposed method compared with the state-of-art methods.
ISSN:1751-9632
1751-9640
1751-9640
DOI:10.1049/iet-cvi.2017.0487