Face Retouching

Dataset used in In-group and Out-group Performance Bias in Facial Retouching Detection paper.

To analyze the effects of retouching on faces from different demographics and test performance of retouching detection algorithms, The Multi-Demographic Retouched Faces (MDRF) database is being shared. This includes original and retouched face images of 600 subjects from 3 origins and 2 genders. This also includes human responses to the task of facial retouching detection. Dataset can be downloaded upon signing an agreement from https://cvrl.nd.edu/projects/data/

Please cite the following when you use the dataset:

@inproceedings{bharati2022,

title={In-group and out-group performance bias in facial retouching detection},

author={Bharati, Aparna and Vatsa, Mayank and Singh, Richa and Bowyer, Kevin W},

booktitle={2022 IEEE international joint conference on biometrics (IJCB)},

year={2022},

organization={IEEE}

}