In this paper, we address the problem of apparent age estimation. Differentfrom estimating the real age of individuals, in which each face image has asingle age label, in this problem, face images have multiple age labels,corresponding to the ages perceived by the annotators, when they look at theseimages. This provides an intriguing computer vision problem, since in genericimage or object classification tasks, it is typical to have a single groundtruth label per class. To account for multiple labels per image, instead ofusing average age of the annotated face image as the class label, we havegrouped the face images that are within a specified age range. Using these agegroups and their age-shifted groupings, we have trained an ensemble of deeplearning models. Before feeding an input face image to a deep learning model,five facial landmark points are detected and used for 2-D alignment. We haveemployed and fine tuned convolutional neural networks (CNNs) that are based onVGG-16 [24] architecture and pretrained on the IMDB-WIKI dataset [22]. Theoutputs of these deep learning models are then combined to produce the finalestimation. Proposed method achieves 0.3668 error in the final ChaLearn LAP2016 challenge test set [5].
展开▼