Unveiling the stories written on faces, this decoder unlocks the secrets of age and gender, bridging the gap between appearance and identity.
Dataset: UTKFace
The dataset has over 20 thousand sample of faces with annotations of gender, age and ethnicity. I have used a custom dataset in this example. The dataset still performs poorly on the unseen data. So, I am working on that part. Also, trying to use the existing CNN architecture like resnet50, vgg16.