Deep Faces: Advancing Age and Gender Classification using Facial Images with Deep Features

Ammad Noor; Wakeel Ahmad; Syed M. Adnan Shah

Authors

Ammad Noor Department of Computer Science, University of Engineering and Technology Taxila, Pakistan.
Wakeel Ahmad Department of Computer Science, University of Engineering and Technology Taxila, Pakistan.
Syed M. Adnan Shah Department of Computer Science, University of Engineering and Technology Taxila, Pakistan.

Keywords:

Age and Gender Classification, Deep Features, VGG-19, CNN, Identity Recognition, Human Facial Features

Abstract

In the realm of identity recognition and social interactions, human facial features play a pivotal role. Accurate age estimation and gender classification from facial images have practical implications across various fields, including biometrics, surveillance, and personalized services. This study presents a novel approach that harnesses deep features extracted by the VGG-19 architecture for age and gender prediction, employing a custom convolutional neural network (CNN) for classification. Leveraging the UTKFace dataset, encompassing a diverse collection of facial images with annotated age and gender labels spanning various ages, ethnicities, and gender representations, provides a robust foundation for model training and evaluation. Deep features extracted from the VGG-19 architecture serve as rich representations of facial patterns, enabling our model to discern discriminative cues for age and gender. These deep features are input to CNN model, which is fine-tuned specifically for age and gender classification. The model comprises input layer, Dense layers, incorporating dropout and batch normalization to mitigate overfitting, and Activation Functions Sigmoid for gender classification and SoftMax for Age group classification. The dataset is divided into training and validation sets (70% and 30%, respectively), enabling the model to learn to map VGG-19 features to age and gender labels. To evaluate the performance of the model, metrics like accuracy, precision, recall, and F1-score are employed. The proposed model achieves an impressive 78.67% accuracy in predicting age and 97.02% accuracy in gender classification on the UTKFace dataset, outperforming traditional methods despite challenges posed by variations in lighting, pose, and expression. The robustness of our approach is evidenced by its capability to handle diverse gender representations.

References

R. Angulu, J. R. Tapamo, A. O. J. E. J. o. I. Adewumi, and V. Processing, "Age estimation via face images: a survey," vol. 2018, no. 1, pp. 1-35, 2018.

P. Terhörst, N. Damer, F. Kirchbuchner, and A. Kuijper, "Suppressing gender and age in face templates using incremental variable elimination," in 2019 International Conference on Biometrics (ICB), 2019, pp. 1-8: IEEE.

A. Althnian, N. Aloboud, N. Alkharashi, F. Alduwaish, M. Alrshoud, and H. J. A. S. Kurdi, "Face gender recognition in the wild: an extensive performance comparison of deep-learned, hand-crafted, and fused features with deep and traditional models," vol. 11, no. 1, p. 89, 2020.

X. Tan and B. J. I. t. o. i. p. Triggs, "Enhanced local texture feature sets for face recognition under difficult lighting conditions," vol. 19, no. 6, pp. 1635-1650, 2010.

B. B. Mamatkulovich and H. A. J. E. S. H. Alijon o’g’li, "Facial Image-Based Gender and Age Estimation," vol. 18, pp. 47-50, 2023.

T. P. Kancharlapalli and P. Dwivedi, "A Novel Approach for Age and Gender Detection using Deep Convolution Neural Network," in 2023 10th International Conference on Computing for Sustainable Global Development (INDIACom), 2023, pp. 873-878: IEEE.

V. Raman, K. ELKarazle, P. J. C. S. S. Then, and Engineering, "Artificially Generated Facial Images for Gender Classification Using Deep Learning," vol. 44, no. 2, pp. 1341-1355, 2023.

N. Sharma, R. Sharma, and N. J. W. P. C. Jindal, "Face-based age and gender estimation using improved convolutional neural network approach," vol. 124, no. 4, pp. 3035-3054, 2022.

Y. H. Kim, S. H. Nam, S. B. Hong, and K. R. J. E. S. w. A. Park, "GRA-GAN: Generative adversarial network for image style transfer of Gender, Race, and age," vol. 198, p. 116792, 2022.

S. Kumar, S. Singh, J. Kumar, K. J. M. T. Prasad, and Applications, "Age and gender classification using Seg-Net based architecture and machine learning," vol. 81, no. 29, pp. 42285-42308, 2022.

M. K. Benkaddour, S. Lahlali, and M. Trabelsi, "Human age and gender classification using convolutional neural network," in 2020 2nd international workshop on human-centric smart environments for health and well-being (IHSH), 2021, pp. 215-220: IEEE.

C.-Y. Hsu, L.-E. Lin, C. H. J. M. T. Lin, and Applications, "Age and gender recognition with random occluded data augmentation on facial images," vol. 80, pp. 11631-11653, 2021.

J. Rwigema, J. Mfitumukiza, K. J. B. S. P. Tae-Yong, and Control, "A hybrid approach of neural networks for age and gender classification through decision fusion," vol. 66, p. 102459, 2021.

O. Agbo-Ajala and S. Viriri, "Face-based age and gender classification using deep learning model," in Image and Video Technology: PSIVT 2019 International Workshops, Sydney, NSW, Australia, November 18–22, 2019, Revised Selected Papers 9, 2020, pp. 125-137: Springer.

Hosseini, S., et al. Age and gender classification using wide convolutional neural network and Gabor filter. in 2018 International Workshop on Advanced Image Technology (IWAIT). 2018. IEEE.

Das, A., A. Dantcheva, and F. Bremond. Mitigating bias in gender, age and ethnicity classification: a multi-task convolution neural network approach. in Proceedings of the european conference on computer vision (eccv) workshops. 2018.

Hassan, K.R. and I.H. Ali. Age and Gender Classification using Multiple Convolutional Neural Network. in IOP Conference Series: Materials Science and Engineering. 2020. IOP Publishing.

Khan, K., et al., A multi-task framework for facial attributes classification through end-to-end face parsing and deep convolutional neural networks. 2020. 20(2): p. 328.

Duan, M., et al., A hybrid deep learning CNN–ELM for age and gender classification. 2018. 275: p. 448-461.

Singh, A., V.K.J.N.C. Singh, and Applications, A hybrid transformer–sequencer approach for age and gender classification from in-wild facial images. 2024. 36(3): p. 1149-1165.

Mustapha, M.F., et al. Improving the Accuracy of Gender Classification Based on Skin Tone Using Convolutional Neural Network: Transfer Learning (CNN-TL). in International Conference on Robotics, Vision, Signal Processing and Power Applications. 2024. Springer.

Uysal, M. and M.F.J.D.Ü.M.F.M.D. Demiral, Gender and Age Estimation By Image Processing. 15(1): p. 49-59.