Understanding More Types of Social Relationships Using Clothing and Distance Metric Learning

Image-based social relationships classification is an emerging and challenging problem in social media analysis. Practically, social relationships can provide intelligent entities with a deeper understanding of human behavior and emotion. Essentially, as an extension of our former study, we proceed...

Full description

Saved in:
Bibliographic Details
Published in:Neural processing letters 2022-12, Vol.54 (6), p.4951-4970
Main Authors: Qin, Xiaoqian, Liu, Dakun, Wang, Dong, Gui, Bin
Format: Article
Language:English
Subjects:
Annotations
Artificial Intelligence
Classification
Complex Systems
Computational Intelligence
Computer Science
Kinship
Learning
Neural networks
Privacy
Similarity measures
Social networks
Teaching methods
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Image-based social relationships classification is an emerging and challenging problem in social media analysis. Practically, social relationships can provide intelligent entities with a deeper understanding of human behavior and emotion. Essentially, as an extension of our former study, we proceed with the following work. First, an extended social relationships database named SRI2.0 is introduced based on the coverage of social relationships categories. Second, we propose to exploit clothing information to extract the relevance of clothing characteristics among the group in images. Finally, we propose a metric learning method by using the correlation similarity measure to better highlight the differences between images with different types of social relationships. Experimental results demonstrate the efficiency of our proposed methods.
ISSN:1370-4621
1573-773X
DOI:10.1007/s11063-022-10842-5