Enhanced Age, Gender, Race Estimation Using Multi-task CNN
Article Sidebar
Main Article Content
Abstract: This research on Enhanced Age, Gender, and Race Estimation Using multi-task CNN presents a comprehensive evaluation of a multi-task deep convolutional neural network (CNN) model designed to simultaneously estimate age, gender, and race from facial images. The testing utilizes real-world datasets, such as UTKFace, Adience, and MORPH II, along with a synthetic dataset that simulates ideal conditions (100% prediction accuracy) for baseline validation. The evaluation includes Mean Absolute Error (MAE) for age estimation, classification accuracy for gender and race, and one-off age accuracy to account for predictions in neighboring classes. Confusion matrices and distribution analysis provide deeper insights into the model's performance across different demographic groups. Although datasets such as UTKFace, Adience, and MORPH II present challenges due to variations in age, gender, and distributions, the proposed model demonstrates strong and high predictive accuracy. The results show that the proposed model surpasses state-of-the-art approaches, achieving an age estimation MAE of 2.95, gender classification accuracy of 98.3%, race classification accuracy of 93.1%, and one-off age accuracy of 90.7%. The addition of synthetic data proved beneficial in enhancing model robustness by mitigating demographic bias and improving prediction reliability. The findings of this study have practical implications for developing fair and reliable demographic estimation systems, with potential applications in security, human-computer interaction, and healthcare. Future work will focus on integrating attention mechanisms, fairness-aware learning, and domain adaptation techniques to enhance accuracy across diverse populations and uncontrolled environments.
Downloads
References
M. M. Iqbal, A. Rukhsar, and S. K. Baliarsingh, ‘A CNN-based Prediction Model for Age, Gender, and Ethnicity Using Facial Images’, in 2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT), Delhi, India: IEEE, July 2023, pp. 1–7. doi: 10.1109/icccnt56998.2023.10306826.
K. Maag and A. Fischer, ‘Uncertainty-weighted Loss Functions for Improved Adversarial Attacks on Semantic Segmentation’, Oct. 26, 2023, arXiv: arXiv:2310.17436. doi: 10.48550/arXiv.2310.17436.
V. Rajpoot, R. Agarwal, J. Agarwal, and R. Pradhan, ‘A study on predicting age using Convolution Neural Network CNN’, in 2021 5th International Conference on Information Systems and Computer Networks (ISCON), Mathura, India: IEEE, Oct. 2021, pp. 1–6. doi: 10.1109/ISCON52037.2021.9702351.
Y. Zhang et al., ‘Attention-based 3D CNN with Multi-layer Features for Alzheimer’s Disease Diagnosis using Brain Images’, Aug. 10, 2023, arXiv: arXiv:2308.05655. doi: 10.48550/arXiv.2308.05655.
K. Simonyan and A. Zisserman, ‘Very Deep Convolutional Networks for Large-Scale Image Recognition’, Sept. 04, 2014, arXiv: arXiv:1409.1556. doi: 10.48550/arXiv.1409.1556.
L. Alzubaidi et al., ‘Review of deep learning: concepts, CNN architectures, challenges, applications, future directions’, J. Big Data, vol. 8, no. 1, p. 53, Mar. 2021, doi: 10.1186/s40537-021-00444-8.
S. Park, S.-H. Park, J.-S. Oh, Y.-K. Noh, J. K. Hur, and J.-W. Nam, ‘shRNAI: a deep neural network for the design of highly potent shRNAs’, Jan. 09, 2024, bioRxiv. Doi: 10.1101/2024.01.09.574789.
Q.-L. Z. Y.-B. Yang, ‘SA-Net: Shuffle Attention for Deep Convolutional Neural Networks’, Jan. 30, 2021, arXiv: arXiv:2102.00240. doi: 10.48550/arXiv.2102.00240.
P. Foggia, A. Greco, A. Roberto, A. Saggese, and M. Vento, ‘Identity, Gender, Age, and Emotion Recognition from Speaker Voice with Multi-task Deep Networks for Cognitive Robotics’, Cogn. Comput., vol. 16, no. 5, pp. 2713–2723, Sept. 2024, doi: 10.1007/s12559-023-10241-5.
Y. Lin et al., ‘Multi-task deep convolutional neural network for weed detection and navigation path extraction’, Comput. Electron. Agric., vol. 229, p. 109776, Feb. 2025, doi: 10.1016/j.compag.2024.109776.
A. Saroop, P. Ghugare, S. Mathamsetty, and V. Vasani, ‘Facial Emotion Recognition: A multi-task approach using deep learning’, Oct. 28, 2021, arXiv: arXiv:2110.15028. doi: 10.48550/arXiv.2110.15028.
S. Siddique, M. A. Haque, R. George, K. D. Gupta, D. Gupta, and M. J. H. Faruk, ‘Survey on Machine Learning Biases and Mitigation Techniques’, Digital, vol. 4, no. 1, pp. 1–68, Mar. 2024, doi: 10.3390/digital4010001.
A. Puc, V. Štruc, and K. Grm, ‘Analysis of Race and Gender Bias in Deep Age Estimation Models’, in 2020 28th European Signal Processing Conference (EUSIPCO), Jan. 2021, pp. 830–834. doi: 10.23919/Eusipco47968.2020.9287219.
M. I. Zaman and N. Ahmed, ‘Deep Learning-Based Age Estimation and Gender Deep Learning-Based Age Estimation and Gender Classification for Targeted Advertisement’, July 24, 2025, arXiv: arXiv:2507.18565. doi: 10.48550/arXiv.2507.18565.
S. Subramanian, B. Tseng, R. Barbieri, and E. N. Brown, ‘An unsupervised automated paradigm for artifact removal from electrodermal activity in an uncontrolled clinical setting’, Physiol. Meas., vol. 43, no. 11, p. 115005, Nov. 2022, doi: 10.1088/1361-6579/ac92bd.
B. Zhang and Y. Bao, ‘Age Estimation of Faces in Videos Using Head Pose Estimation and Convolutional Neural Networks’, Sensors, vol. 22, no. 11, p. 4171, May 2022, doi: 10.3390/s22114171.
V. Raman, K. ELKarazle, and P. Then, ‘Gender-specific Facial Age Group Classification Using Deep Learning’, Intell. Autom. Soft Comput., vol. 34, no. 1, pp. 105–118, 2022, doi: 10.32604/iasc 2022.025608.
X. Du, Y. Sun, Y. Song, H. Sun, and L. Yang, ‘A Comparative Study of Different CNN Models and Transfer Learning Effect for Underwater Object Classification in Side-Scan Sonar Images’, Remote Sens., vol. 15, no. 3, Art. no. 3, Jan. 2023, doi: 10.3390/rs15030593.
K. Sampath, S. Rajagopal, and A. Chintanpalli, ‘A comparative analysis of CNN-based deep learning architectures for early diagnosis of bone cancer using CT images’, Sci. Rep., vol. 14, p. 2144, Jan. 2024, doi: 10.1038/s41598-024-52719-8.
Md. N. Islam Opu, T. K. Koly, A. Das, and A. Dey, ‘A Lightweight Deep Convolutional Neural Network Model for Real-Time Age and Gender Prediction’, in 2020 Third International Conference on Advances in Electronics, Computers and Communications (ICAECC), Bengaluru, India: IEEE, Dec. 2020, pp. 1–6. doi: 10.1109/icaecc50550.2020.9339503.
S.-J. Park et al., ‘Automatic and robust estimation of sex and chronological age from panoramic radiographs using a multi-task deep learning network: a study on a South Korean population’, Int. J. Legal Med., vol. 138, no. 4, pp. 1741–1757, July 2024, doi: 10.1007/s00414-024-03204-4.
Y. Cai, X. Li, Y. Zhang, J. Li, F. Zhu, and L. Rao, ‘Multimodal sentiment analysis based on multi-layer feature fusion and multi-task learning’, Sci. Rep., vol. 15, no. 1, p. 2126, Jan. 2025, doi: 10.1038/s41598-025-85859-6.
A. Puc, V. Struc, and K. Grm, ‘Analysis of Race and Gender Bias in Deep Age Estimation Models’, in 2020 28th European Signal Processing Conference (EUSIPCO), Amsterdam, Netherlands: IEEE, Jan. 2021, pp. 830–834. doi: 10.23919/eusipco47968.2020.9287219.
M. K. Benkaddour, ‘CNN-Based Features Extraction for Age Estimation and Gender Classification’, Informatica, vol. 45, no. 5, Art. no. 5, Aug. 2021, doi: 10.31449/inf.v45i5.3262.
O. Oak, R. Nazre, S. Naigaonkar, S. Sawant, and H. Vaidya, ‘A Comparative Analysis of CNN-based Deep Learning Models for Landslide Detection’, in 2024 Asian Conference on Intelligent Technologies (ACOIT), Sept. 2024, pp. 1–6. doi: 10.1109/ACOIT62457.2024.10939989.
F. Anderson, A. Carson, L. Whitehead, and K. Burau, ‘Age, Race and Gender Spatiotemporal Disparities of COPD Emergency Room Visits in Houston, Texas’, Occup. Dis. Environ. Med., vol. 3, no. 1, Art. no. 1, Feb. 2015, doi: 10.4236/odem.2015.31001.
M. Trigka and E. Dritsas, ‘A Comprehensive Survey of Deep Learning Approaches in Image Processing’, Sensors, vol. 25, no. 2, Art. no. 2, Jan. 2025, doi: 10.3390/s25020531.
‘Multi-task learning on the edge for effective gender, age, ethnicity, and emotion recognition’, Eng. Appl. Artif. Intell., vol. 118, p. 105651, Feb. 2023, doi: 10.1016/j.engappai.2022.105651.
S. Sakib, K. Deb, P. K. Dhar, and O.-J. Kwon, ‘A Framework for Pedestrian Attribute Recognition Using Deep Learning’, Appl. Sci., vol. 12, no. 2, p. 622, Jan. 2022, doi: 10.3390/app12020622.
N. Sarafianos, X. Xu, and I. A. Kakadiaris, ‘Deep Imbalanced Attribute Classification using Visual Attention Aggregation’, presented at the Proceedings of the European Conference on Computer Vision (ECCV), 2018, pp. 680–697. Accessed: Sept. 04, 2025. [Online]. Available: https://openaccess.thecvf.com/content_ECCV_2018/html/Nikolaos_Sarafianos_Deep_Imbalanced_Attribute_ECCV_2018_paper.html
H. Mahamivanan et al., ‘Material recognition for construction quality monitoring using deep learning methods’, Constr. Innov., vol. 25, no. 3, pp. 732–760, Apr. 2025, doi: 10.1108/CI-04-2022-0074.
S. W. Shende, J. V. Tembhurne, and N. A. Ansari, ‘Deep learning based authentication schemes for smart devices in different modalities: progress, challenges, performance, datasets and future directions’, Multimed. Tools Appl., vol. 83, no. 28, pp. 71451–71493, Aug. 2024, doi: 10.1007/s11042-024-18350-5.
C. Wen, X. Zhang, X. Yao, and J. Yang, ‘Ordinal Label Distribution Learning’, presented at the Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023, pp. 23481–23491. Accessed: Sept. 04, 2025. [Online]. Available: https://openaccess.thecvf.com/content/ICCV2023/html/Wen_Ordinal_Label_Distribution_Learning_ICCV_2023_paper.html
J.-C. Xie and C.-M. Pun, ‘Deep and Ordinal Ensemble Learning for Human Age Estimation From Facial Images’, IEEE Trans. Inf. Forensics Secur., vol. 15, pp. 2361–2374, 2020, doi: 10.1109/TIFS.2020.2965298.
X. Zhou, Z. Wei, M. Xu, S. Qu, and G. Guo, ‘Facial Depression Recognition by Deep Joint Label Distribution and Metric Learning’, IEEE Trans. Affect. Comput., vol. 13, no. 3, pp. 1605–1618, July 2022, doi: 10.1109/TAFFC.2020.3022732.
Z. Bao, Z. Tan, J. Wan, X. Ma, G. Guo, and Z. Lei, ‘Divergence-Driven Consistency Training for Semi-Supervised Facial Age Estimation’, IEEE Trans. Inf. Forensics Secur., vol. 18, pp. 221–232, 2023, doi: 10.1109/TIFS.2022.3218431.
B. Ghojogh and A. Ghodsi, ‘Attention Mechanism, Transformers, BERT, and GPT: Tutorial and Survey’, Dec. 2020. Accessed: Sept. 04, 2025. [Online]. Available: https://hal.science/hal-04637647
A. M. Hafiz, S. A. Parah, and R. U. A. Bhat, ‘Attention mechanisms and deep learning for machine vision: A survey of the state of the art’, June 03, 2021, arXiv: arXiv:2106.07550. doi: 10.48550/arXiv.2106.07550.
C. Szegedy, S. Ioffe, V. Vanhoucke, and A. Alemi, ‘Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning’, Proc. AAAI Conf. Artif. Intell., vol. 31, no. 1, Feb. 2017, doi: 10.1609/aaai.v31i1.11231.
X. Zhang, R. Jiang, W. Gao, R. Willett, and M. Maire, ‘Residual Connections Harm Generative Representation Learning’, Jan. 31, 2025, arXiv: arXiv:2404.10947. doi: 10.48550/arXiv.2404.10947.
K. He, X. Zhang, S. Ren, and J. Sun, ‘Identity Mappings in Deep Residual Networks’, in Computer Vision – ECCV 2016, B. Leibe, J. Matas, N. Sebe, and M. Welling, Eds, Cham: Springer International Publishing, 2016, pp. 630–645. doi: 10.1007/978-3-319-46493-0_38.
M. Scholkemper, X. Wu, A. Jadbabaie, and M. T. Schaub, ‘Residual Connections and Normalization Can Provably Prevent Oversmoothing in GNNs’, May 26, 2025, arXiv: arXiv:2406.02997. doi: 10.48550/arXiv.2406.02997.
M. Crawshaw, ‘Multi-Task Learning with Deep Neural Networks: A Survey’, Sept. 10, 2020, arXiv: arXiv:2009.09796. doi: 10.48550/arXiv.2009.09796.
S. Vandenhende, S. Georgoulis, W. Van Gansbeke, M. Proesmans, D. Dai, and L. Van Gool, ‘Multi-Task Learning for Dense Prediction Tasks: A Survey’, IEEE Trans. Pattern Anal. Mach. Intell., vol. 44, no. 7, pp. 3614–3633, July 2022, doi: 10.1109/TPAMI.2021.3054719.
G. Hu et al., ‘Deep Multi-Task Learning to Recognise Subtle Facial Expressions of Mental States’, presented at the Proceedings of the European Conference on Computer Vision (ECCV), 2018, pp. 103–119. Accessed: Sept. 04, 2025. [Online]. Available: https://openaccess.thecvf.com/content_ECCV_2018/html/Guosheng_Hu_Deep_Multi-Task_Learning_ECCV_2018_paper.html
Y. Zhang and Q. Yang, ‘A Survey on Multi-Task Learning’, IEEE Trans. Knowl. Data Eng., vol. 34, no. 12, pp. 5586–5609, Dec. 2022, doi: 10.1109/TKDE.2021.3070203.
S. Naaz, H. Pandey, and C. Lakshmi, ‘Deep Learning based age and gender detection using facial images’, in 2024 International Conference on Advances in Computing, Communication and Applied Informatics (ACCAI), May 2024, pp. 1–11. doi: 10.1109/ACCAI61061.2024.10601975.
M. S. Islam and M. S. Mia, ‘Human Gender and Age Detection Through Image Processing Technique’, Feb. 2025, Accessed: Sept. 04, 2025. [Online]. Available: http://ar.cou.ac.bd:8080/xmlui/handle/123456789/109
V. Vigilante, ‘Intelligent embedded system for facial soft biometrics in social robotics’, Doctoral Thesis, Università degli Studi di Salerno, 2021. doi: 10.14273/unisa-4336.
A. Garain, B. Ray, P. K. Singh, A. Ahmadian, N. Senu, and R. Sarkar, ‘GRA_Net: A Deep Learning Model for Classification of Age and Gender From Facial Images’, IEEE Access, vol. 9, pp. 85672–85689, 2021, doi: 10.1109/ACCESS.2021.3085971.
K. Kotwal and S. Marcel, ‘Review of Demographic Fairness in Face Recognition’, IEEE Trans. Biom. Behav. Identity Sci., pp. 1–1, 2025, doi: 10.1109/TBIOM.2025.3601217.
T. P. Pagano et al., ‘Bias and Unfairness in Machine Learning Models: A Systematic Review on Datasets, Tools, Fairness Metrics, and Identification and Mitigation Methods’, Big Data Cogn. Comput., vol. 7, no. 1, p. 15, Mar. 2023, doi: 10.3390/bdcc7010015.
P. Chen, L. Wu, and L. Wang, ‘AI Fairness in Data Management and Analytics: A Review on Challenges, Methodologies and Applications’, Appl. Sci., vol. 13, no. 18, p. 10258, Jan. 2023, doi: 10.3390/app131810258.
S. V. Chinta et al., ‘FairAIED: Navigating Fairness, Bias, and Ethics in Educational AI Applications’, July 26, 2024, arXiv: arXiv:2407.18745. doi: 10.48550/arXiv.2407.18745.
T. P. Pagano et al., ‘Bias and Unfairness in Machine Learning Models: A Systematic Review on Datasets, Tools, Fairness Metrics, and Identification and Mitigation Methods’, Big Data Cogn. Comput., vol. 7, no. 1, p. 15, Mar. 2023, doi: 10.3390/bdcc7010015.
J. Jiang et al., ‘ForkMerge: Mitigating Negative Transfer in Auxiliary-Task Learning’, Adv. Neural Inf. Process. Syst., vol. 36, pp. 30367–30389, Dec. 2023.
S. Liu, Y. Liang, and A. Gitter, ‘Loss-Balanced Task Weighting to Reduce Negative Transfer in Multi-Task Learning’, Proc. AAAI Conf. Artif. Intell., vol. 33, no. 01, pp. 9977–9978, July 2019, doi: 10.1609/aaai.v33i01.33019977.
S. Wu, H. R. Zhang, and C. Ré, ‘Understanding and Improving Information Transfer in Multi-Task Learning’, May 02, 2020, arXiv: arXiv:2005.00944. doi: 10.48550/arXiv.2005.00944.
Z. Meng, X. Yao, and L. Sun, ‘Multi-Task Distillation: Towards Mitigating the Negative Transfer in Multi-Task Learning’, in 2021 IEEE International Conference on Image Processing (ICIP), Sept. 2021, pp. 389–393. doi: 10.1109/ICIP42928.2021.9506618.
M. F. Senussi, M. Abdalla, M. S. Kasem, M. Mahmoud, B. Yagoub, and H.-S. Kang, ‘A Comprehensive Review on Light Field Occlusion Removal: Trends, Challenges, and Future Directions’, IEEE Access, vol. 13, pp. 42472–42493, 2025, doi: 10.1109/ACCESS.2025.3548133.
M. De Marsico, M. Nappi, D. Riccio, and H. Wechsler, ‘Robust Face Recognition for Uncontrolled Pose and Illumination Changes’, IEEE Trans. Syst. Man Cybern. Syst., vol. 43, no. 1, pp. 149–163, Jan. 2013, doi: 10.1109/TSMCA.2012.2192427.
M. Cormier, A. Clepe, A. Specker, and J. Beyerer, ‘Where Are We With Human Pose Estimation in Real-World Surveillance?’, presented at the Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2022, pp. 591–601. Accessed: Sept. 04, 2025. [Online]. Available: https://openaccess.thecvf.com/content/WACV2022W/RWS/html/Cormier_Where_Are_We_With_Human_Pose_Estimation_in_Real-World_Surveillance_WACVW_2022_paper.html
R. Shrestha, K. Kafle, and C. Kanan, ‘An Investigation of Critical Issues in Bias Mitigation Techniques’, presented at the Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2022, pp. 1943–1954. Accessed: Sept. 04, 2025. [Online]. Available: https://openaccess.thecvf.com/content/WACV2022/html/Shrestha_An_Investigation_of_Critical_Issues_in_Bias_Mitigation_Techniques_WACV_2022_paper.html
ChenZhenpeng, Z. M, SarroFederica, and HarmanMark, ‘A Comprehensive Empirical Study of Bias Mitigation Methods for Machine Learning Classifiers’, ACM Trans. Softw. Eng. Methodol., May 2023, doi: 10.1145/3583561.
S. A. Manavski, ‘CUDA Compatible GPU as an Efficient Hardware Accelerator for AES Cryptography’, in 2007 IEEE International Conference on Signal Processing and Communications, Nov. 2007, pp. 65–68. doi: 10.1109/ICSPC.2007.4728256.
A. Puc, V. Struc, and K. Grm, ‘Analysis of Race and Gender Bias in Deep Age Estimation Models’, in 2020 28th European Signal Processing Conference (EUSIPCO), Amsterdam, Netherlands: IEEE, Jan. 2021, pp. 830–834. doi: 10.23919/eusipco47968.2020.9287219.
C. H. Nga, K.-T. Nguyen, N. C., Tran, and J.-C. Wang, ‘Transfer Learning for Gender and Age Prediction’, in 2020 IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan), Taoyuan, Taiwan: IEEE, Sept. 2020, pp. 1–2. doi: 10.1109/ICCE-Taiwan49838.2020.9258347.
M. K. Benkaddour, ‘CNN-Based Features Extraction for Age Estimation and Gender Classification’, Informatica, vol. 45, no. 5, Art. no. 5, Aug. 2021, doi: 10.31449/inf.v45i5.3262.
H. Liao, L. Yuan, M. Wu, L. Zhong, G. Jin, and N. Xiong, ‘Face Gender and Age Classification Based on Multi-Task, Multi-Instance and Multi-Scale Learning’, Appl. Sci., vol. 12, no. 23, p. 12432, Jan. 2022, doi: 10.3390/app122312432.
J. Wang and X. Geng, ‘Label Distribution Learning Machine’, in Proceedings of the 38th International Conference on Machine Learning, PMLR, July 2021, pp. 10749–10759. Accessed: July 09, 2025. [Online]. Available: https://proceedings.mlr.press/v139/wang21h.html
H. Han, A. K. Jain, F. Wang, S. Shan, and X. Chen, ‘Heterogeneous Face Attribute Estimation: A Deep Multi-Task Learning Approach’, IEEE Trans. Pattern Anal. Mach. Intell., vol. 40, no. 11, pp. 2597–2609, Nov. 2018, doi: 10.1109/TPAMI.2017.2738004.
M. M. Hosseini, A. P. Fard, and M. H. Mahoor, ‘Faces of Fairness: Examining Bias in Facial Expression Recognition Datasets and Models’, Feb. 16, 2025, arXiv: arXiv:2502.11049. doi: 10.48550/arXiv.2502.11049.
D. Bontempi et al., ‘FaceAge, a deep learning system to estimate biological age from face photographs to improve prognostication: a model development and validation study’, Lancet Digit. Health, vol. 7, no. 6, June 2025, doi: 10.1016/j.landig.2025.03.002.
M. Rohani, H. Farsi, and S. Mohamadzadeh, ‘Facial Feature Recognition with Multi-task Learning and Attention-based Enhancements’, Iran. J. Energy Environ., vol. 16, no. 1, pp. 136–144, Jan. 2025, doi: 10.5829/ijee 2025.16.01.14.
This work is licensed under a Creative Commons Attribution 4.0 International License.
All articles published in our journal are licensed under CC-BY 4.0, which permits authors to retain copyright of their work. This license allows for unrestricted use, sharing, and reproduction of the articles, provided that proper credit is given to the original authors and the source.