Significance of Softmax-Based Features in Comparison to Distance Metric Learning-Based Features

End-to-end distance metric learning (DML) has been applied to obtain features useful in many computer vision tasks. However, these DML studies have not provided equitable comparisons between features extracted from DML-based networks and softmax-based networks. In this paper, we present objective co...

Full description

Saved in:
Bibliographic Details
Published inIEEE transactions on pattern analysis and machine intelligence Vol. 42; no. 5; pp. 1279 - 1285
Main Authors Horiguchi, Shota, Ikami, Daiki, Aizawa, Kiyoharu
Format Journal Article
LanguageEnglish
Published United States IEEE 01.05.2020
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects
Animals
Automobiles
Birds
classification
Cluster Analysis
Computer vision
Databases, Factual
Deep Learning
Dimensionality reduction
distance metric learning
Feature extraction
Image Processing, Computer-Assisted - methods
Learning
Measurement
Network architecture
Neural Networks, Computer
Principal component analysis
retrieval
Task analysis
Online AccessGet full text

Cover

More Information
Summary:End-to-end distance metric learning (DML) has been applied to obtain features useful in many computer vision tasks. However, these DML studies have not provided equitable comparisons between features extracted from DML-based networks and softmax-based networks. In this paper, we present objective comparisons between these two approaches under the same network architecture.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0162-8828
1939-3539
2160-9292
DOI:10.1109/TPAMI.2019.2911075