#Age Estimation and Gender Classification

Dataset downloading You will use a subset of the UTKFace dataset. A train val folder has been created by choosing a subset (5,000 face images) from the UTKFace dataset for you to train and validate your model. It is a shared google drive folder below. https://drive.google.com/drive/folders/1CJDxAAb930hInvOGt_J3TsPPjao3pgwU?usp=sharing

Click the link above. It must be in the “shared with me” in your Google Drive. • Right-click this folder, choose “Add shortcut to Drive”, then specify the path where you would like to put it. • Then you can access its path after you mount the drive.

UTKFace dataset is a large-scale face dataset with long age span (range from 0 to 116 years old). The dataset consists of over 20,000 face images with annota- tions of age, gender, and ethnicity. The images cover large variation in pose, fa- cial expression, illumination, occlusion, resolution, etc. The labels of each image are embedded in the file name, formatted as [age] [gender] [race] [date&time].jpg.

Ethical disclaimer: We acknowledge that some people might feel uncom- fortable about using this dataset, such as predicting a person’s age and gender purely based on his/her appearance, only female and male included in the gen- der label, etc. We don’t consider moral issues simply because we only use it for a ML exercise for this course. However, you MUST consider potential moral issues when using any dataset in your future research or career.