Facial Age Estimation using Convolutional Neural Networks
This is an incremental student project addressing age estimation from facial images, with limited practical impact due to inferior benchmark performance.
The paper tackled facial age estimation by training a convolutional neural network from scratch on three datasets, achieving 52% categorical accuracy on a test set but performing poorly on the Adience benchmark with 30% exact accuracy.
This paper is a part of a student project in Machine Learning at the Norwegian University of Science and Technology. In this paper, a deep convolutional neural network with five convolutional layers and three fully-connected layers is presented to estimate the ages of individuals based on images. The model is in its entirety trained from scratch, where a combination of three different datasets is used as training data. These datasets are the APPA dataset, UTK dataset, and the IMDB dataset. The images were preprocessed using a proprietary face-recognition software. Our model is evaluated on both a held-out test set, and on the Adience benchmark. On the test set, our model achieves a categorical accuracy of 52%. On the Adience benchmark, our model proves inferior compared with other leading models, with an exact accuray of 30%, and an one-off accuracy of 46%. Furthermore, a script was created, allowing users to estimate their age directly using their web camera. The script, alongside all other code, is located in our GitHub repository: AgeNet.