DeepAge: A Convolutional Neural Network Approach for Accurate Age Estimation

• Published in: 2024 15th International Conference on Computing Communication and Networking Technologies (ICCCNT) pp. 1 - 5
• Main Authors: Bhardwaj, Vivek, Agarwal, Sanchita, Kumar, Mukesh, Kaur, Navjeet, Dhaliwal, Balwinder Kaur, Kishore, Jaydeep
• Format: Conference Proceeding
• Language: English
• Published: IEEE 24-06-2024
• Subjects: Accuracy; Age Estimation; Artificial Intelligence; CNNs; Computer architecture; Convolutional neural networks; Estimation; Human computer interaction; Image Processing; Measurement uncertainty; Optimization; Predictive models; Robustness; Training
• ISSN: 2473-7674
• DOI: 10.1109/ICCCNT61001.2024.10725562

In a variety of applications, including human-computer interaction, targeted marketing, and a project offers a reliable method for precise age estimation based on convolutional neural networks (CNNs). Making use of an extensive dataset, the model applies sophisticated data augmentation methods to improve training robustness and diversity. The incorporation of regularization, techniques like dropout and L2 regularization, which greatly reduce overfitting and enhance generalization to new data, is one of the major advances. The outcomes show a significant improvement in the training and validation measures, with the mean absolute error (MAE) and mean squared error (MSE) for age prediction showing the most reductions. The model performs admirably despite a small amount of overfitting, demonstrating the effectiveness of the used methods. This work highlights CNNs' promise in age estimation problems and provides a strong basis for further improvements, such as the addition of bigger datasets and more complex architectures. This thorough methodology guarantees that the model can be applied in real-world situations, which makes it a significant contribution to the subject.