Selective multi-descriptor fusion for face identification

Research output: Contribution to journalArticle

Abstract

Over the last 2 decades, face identification has been an active field of research in computer vision. As an important class of image representation methods for face identification, fused descriptor-based methods are known to lack sufficient discriminant information, especially when compared with deep learning-based methods. This paper presents a new face representation method, multi-descriptor fusion (MDF), which represents face images through a combination of multiple descriptors, resulting in hyper-high dimensional fused descriptor features. MDF enables excellent performance in face identification, exceeding the state-of-the-art, but it comes with high memory and computational costs. As a solution to the high cost problem, this paper also presents an optimisation method, discriminant ability-based multi-descriptor selection (DAMS), to select a subset of descriptors from the set of 65 initial descriptors whilst maximising the discriminant ability. The MDF face representation, after being refined by DAMS, is named selective multi-descriptor fusion (SMDF). Compared with MDF, SMDF has much smaller feature dimension and is thus usable on an ordinary PC, but still has similar performance. Various experiments are conducted on the CAS-PEAL-R1 and LFW datasets to demonstrate the performance of the proposed methods.
LanguageEnglish
JournalInternational Journal of Machine Learning and Cybernetics
Early online date15 Feb 2019
Publication statusE-pub ahead of print - 15 Feb 2019

Fingerprint

Fusion reactions
Computer vision
Costs
Data storage equipment
Experiments

Cite this

@article{6a1605a860414719a14fd6613a988438,
title = "Selective multi-descriptor fusion for face identification",
abstract = "Over the last 2 decades, face identification has been an active field of research in computer vision. As an important class of image representation methods for face identification, fused descriptor-based methods are known to lack sufficient discriminant information, especially when compared with deep learning-based methods. This paper presents a new face representation method, multi-descriptor fusion (MDF), which represents face images through a combination of multiple descriptors, resulting in hyper-high dimensional fused descriptor features. MDF enables excellent performance in face identification, exceeding the state-of-the-art, but it comes with high memory and computational costs. As a solution to the high cost problem, this paper also presents an optimisation method, discriminant ability-based multi-descriptor selection (DAMS), to select a subset of descriptors from the set of 65 initial descriptors whilst maximising the discriminant ability. The MDF face representation, after being refined by DAMS, is named selective multi-descriptor fusion (SMDF). Compared with MDF, SMDF has much smaller feature dimension and is thus usable on an ordinary PC, but still has similar performance. Various experiments are conducted on the CAS-PEAL-R1 and LFW datasets to demonstrate the performance of the proposed methods.",
author = "X Wei and H. Wang and Bryan Scotney and Huan Wan",
year = "2019",
month = "2",
day = "15",
language = "English",
journal = "International Journal of Machine Learning and Cybernetics",
issn = "1868-8071",

}

TY - JOUR

T1 - Selective multi-descriptor fusion for face identification

AU - Wei, X

AU - Wang, H.

AU - Scotney, Bryan

AU - Wan, Huan

PY - 2019/2/15

Y1 - 2019/2/15

N2 - Over the last 2 decades, face identification has been an active field of research in computer vision. As an important class of image representation methods for face identification, fused descriptor-based methods are known to lack sufficient discriminant information, especially when compared with deep learning-based methods. This paper presents a new face representation method, multi-descriptor fusion (MDF), which represents face images through a combination of multiple descriptors, resulting in hyper-high dimensional fused descriptor features. MDF enables excellent performance in face identification, exceeding the state-of-the-art, but it comes with high memory and computational costs. As a solution to the high cost problem, this paper also presents an optimisation method, discriminant ability-based multi-descriptor selection (DAMS), to select a subset of descriptors from the set of 65 initial descriptors whilst maximising the discriminant ability. The MDF face representation, after being refined by DAMS, is named selective multi-descriptor fusion (SMDF). Compared with MDF, SMDF has much smaller feature dimension and is thus usable on an ordinary PC, but still has similar performance. Various experiments are conducted on the CAS-PEAL-R1 and LFW datasets to demonstrate the performance of the proposed methods.

AB - Over the last 2 decades, face identification has been an active field of research in computer vision. As an important class of image representation methods for face identification, fused descriptor-based methods are known to lack sufficient discriminant information, especially when compared with deep learning-based methods. This paper presents a new face representation method, multi-descriptor fusion (MDF), which represents face images through a combination of multiple descriptors, resulting in hyper-high dimensional fused descriptor features. MDF enables excellent performance in face identification, exceeding the state-of-the-art, but it comes with high memory and computational costs. As a solution to the high cost problem, this paper also presents an optimisation method, discriminant ability-based multi-descriptor selection (DAMS), to select a subset of descriptors from the set of 65 initial descriptors whilst maximising the discriminant ability. The MDF face representation, after being refined by DAMS, is named selective multi-descriptor fusion (SMDF). Compared with MDF, SMDF has much smaller feature dimension and is thus usable on an ordinary PC, but still has similar performance. Various experiments are conducted on the CAS-PEAL-R1 and LFW datasets to demonstrate the performance of the proposed methods.

UR - https://link.springer.com/article/10.1007/s13042-019-00929-2

M3 - Article

JO - International Journal of Machine Learning and Cybernetics

T2 - International Journal of Machine Learning and Cybernetics

JF - International Journal of Machine Learning and Cybernetics

SN - 1868-8071

ER -