Multi-Task CNN Model for Attribute Prediction

This paper proposes a joint multi-task learning algorithm to better predict attributes in images using deep convolutional neural networks (CNN). We consider learning binary semantic attributes through a multi-task CNN model, where each CNN will predict one binary attribute. The multitask learni...

Full description

Saved in:
Bibliographic Details
Main Authors: Abdulnabi, Abrar H., Wang, Gang, Lu, Jiwen, Jia, Kui
Other Authors: School of Electrical and Electronic Engineering
Format: Article
Language:English
Published: 2016
Subjects:
Online Access:https://hdl.handle.net/10356/82925
http://hdl.handle.net/10220/40352
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-82925
record_format dspace
spelling sg-ntu-dr.10356-829252020-03-07T13:57:24Z Multi-Task CNN Model for Attribute Prediction Abdulnabi, Abrar H. Wang, Gang Lu, Jiwen Jia, Kui School of Electrical and Electronic Engineering Deep CNN Latent tasks matrix Multi-task learning Semantic attributes This paper proposes a joint multi-task learning algorithm to better predict attributes in images using deep convolutional neural networks (CNN). We consider learning binary semantic attributes through a multi-task CNN model, where each CNN will predict one binary attribute. The multitask learning allows CNN models to simultaneously share visual knowledge among different attribute categories. Each CNN will generate attribute-specific feature representations, and then we apply multi-task learning on the features to predict their attributes. In our multi-task framework, we propose a method to decompose the overall model’s parameters into a latent task matrix and combination matrix. Furthermore, under-sampled classifiers can leverage shared statistics from other classifiers to improve their performance. Natural grouping of attributes is applied such that attributes in the same group are encouraged to share more knowledge. Meanwhile, attributes in different groups will generally compete with each other, and consequently share less knowledge. We show the effectiveness of our method on two popular attribute datasets. ASTAR (Agency for Sci., Tech. and Research, S’pore) MOE (Min. of Education, S’pore) Accepted version 2016-03-31T07:45:42Z 2019-12-06T15:08:21Z 2016-03-31T07:45:42Z 2019-12-06T15:08:21Z 2015 Journal Article Abdulnabi, A. H., Wang, G., Lu, J., & Jia, K. (2015). Multi-Task CNN Model for Attribute Prediction. IEEE Transactions on Multimedia, 17(11), 1949-1959. 1520-9210 https://hdl.handle.net/10356/82925 http://hdl.handle.net/10220/40352 10.1109/TMM.2015.2477680 en IEEE Transactions on Multimedia © 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The published version is available at: [http://dx.doi.org/10.1109/TMM.2015.2477680]. 11 p. application/pdf
institution Nanyang Technological University
building NTU Library
country Singapore
collection DR-NTU
language English
topic Deep CNN
Latent tasks matrix
Multi-task learning
Semantic attributes
spellingShingle Deep CNN
Latent tasks matrix
Multi-task learning
Semantic attributes
Abdulnabi, Abrar H.
Wang, Gang
Lu, Jiwen
Jia, Kui
Multi-Task CNN Model for Attribute Prediction
description This paper proposes a joint multi-task learning algorithm to better predict attributes in images using deep convolutional neural networks (CNN). We consider learning binary semantic attributes through a multi-task CNN model, where each CNN will predict one binary attribute. The multitask learning allows CNN models to simultaneously share visual knowledge among different attribute categories. Each CNN will generate attribute-specific feature representations, and then we apply multi-task learning on the features to predict their attributes. In our multi-task framework, we propose a method to decompose the overall model’s parameters into a latent task matrix and combination matrix. Furthermore, under-sampled classifiers can leverage shared statistics from other classifiers to improve their performance. Natural grouping of attributes is applied such that attributes in the same group are encouraged to share more knowledge. Meanwhile, attributes in different groups will generally compete with each other, and consequently share less knowledge. We show the effectiveness of our method on two popular attribute datasets.
author2 School of Electrical and Electronic Engineering
author_facet School of Electrical and Electronic Engineering
Abdulnabi, Abrar H.
Wang, Gang
Lu, Jiwen
Jia, Kui
format Article
author Abdulnabi, Abrar H.
Wang, Gang
Lu, Jiwen
Jia, Kui
author_sort Abdulnabi, Abrar H.
title Multi-Task CNN Model for Attribute Prediction
title_short Multi-Task CNN Model for Attribute Prediction
title_full Multi-Task CNN Model for Attribute Prediction
title_fullStr Multi-Task CNN Model for Attribute Prediction
title_full_unstemmed Multi-Task CNN Model for Attribute Prediction
title_sort multi-task cnn model for attribute prediction
publishDate 2016
url https://hdl.handle.net/10356/82925
http://hdl.handle.net/10220/40352
_version_ 1681047104455180288