Celebrities often define fashion trends and standards by the styles they choose. To address a rapidly changing demand, designers need to be able to automatically identify trends. The model uses a large database of celebrity images to find trends and facial attributes among such images.
In this case, the model predicts if a celebrity is: a) Wearing Accessories b) Has Thick Nose or Lips
There are 2 data files and 1 folder provided:
face_attribute_train.csv - Ids of training Images + Labels (Wearing_Accessories + Thick_Lips_Nose)
face_images/face_images - Folder containing all of the .jpg images. Filename of the image is the same as "image_id" (see below).
image_id - Filename of relevant image. Image can be found in the folder "face_images/face_images".
Wearing_Accessories - Whether the celebrity is wearing accessories (only provided for train).
Thick_Lips_Nose - Whether the celebrity has a thick lip or nose (only provided for train).
Dataset Link: https://drive.google.com/drive/folders/13ONvC_ahE2r9z4CWhSR8-Rr03nk88ljg?usp=sharing