TY - JOUR
T1 - Multi-dataset fusion for multi-task learning on face attribute recognition
AU - Lu, Hengjie
AU - Xu, Shugong
AU - Wang, Jiahao
N1 - Publisher Copyright:
© 2023 Elsevier B.V.
PY - 2023/9
Y1 - 2023/9
N2 - The goal of face attribute recognition(FAR) is to recognize the attributes of face images, such as gender, race, etc. Multi-dataset fusion aims to train a network with multiple datasets simultaneously, which has significant advantages in deep learning. For example, in FAR, the attribute labels differ with the datasets. When using multi-dataset fusion on FAR, training data and attribute labels can be increased so that the network performance can be improved and more attributes can be recognized simultaneously. Currently, most related researchers take the multi-task learning approach on FAR, where each attribute is treated as a recognition task. However, the existing multi-task learning framework cannot be applied when using multi-dataset fusion on FAR. That is because multi-task learning requires the training data to contain the labels of all tasks, but no data contains all the attribute labels of multiple datasets. A Multi-Dataset and Multi-Task Framework(MDMTF) that takes knowledge distillation as the core is proposed in this paper to deal with the problem above. Three distillation strategies are designed in the MDMTF according to the characteristics of multi-dataset fusion on FAR. Experimental results on CelebA and MAAD-Face datasets demonstrate the effectiveness of our framework and strategies. Compared to training two networks on both datasets, respectively, that is, single dataset training, our method has significant advantages in accuracy, parameters, and computational complexity.
AB - The goal of face attribute recognition(FAR) is to recognize the attributes of face images, such as gender, race, etc. Multi-dataset fusion aims to train a network with multiple datasets simultaneously, which has significant advantages in deep learning. For example, in FAR, the attribute labels differ with the datasets. When using multi-dataset fusion on FAR, training data and attribute labels can be increased so that the network performance can be improved and more attributes can be recognized simultaneously. Currently, most related researchers take the multi-task learning approach on FAR, where each attribute is treated as a recognition task. However, the existing multi-task learning framework cannot be applied when using multi-dataset fusion on FAR. That is because multi-task learning requires the training data to contain the labels of all tasks, but no data contains all the attribute labels of multiple datasets. A Multi-Dataset and Multi-Task Framework(MDMTF) that takes knowledge distillation as the core is proposed in this paper to deal with the problem above. Three distillation strategies are designed in the MDMTF according to the characteristics of multi-dataset fusion on FAR. Experimental results on CelebA and MAAD-Face datasets demonstrate the effectiveness of our framework and strategies. Compared to training two networks on both datasets, respectively, that is, single dataset training, our method has significant advantages in accuracy, parameters, and computational complexity.
KW - Deep learning
KW - Face attribute recognition
KW - Knowledge distillation
KW - Multi-dataset fusion
KW - Multi-task learning
UR - http://www.scopus.com/inward/record.url?scp=85167997292&partnerID=8YFLogxK
U2 - 10.1016/j.patrec.2023.07.015
DO - 10.1016/j.patrec.2023.07.015
M3 - Article
AN - SCOPUS:85167997292
SN - 0167-8655
VL - 173
SP - 72
EP - 78
JO - Pattern Recognition Letters
JF - Pattern Recognition Letters
ER -