Deep learning-based image and video processing is a
tool for image/video understanding, animation, and artistic expression. Similar to most software in this domain,
our work could be used to produce offensive results. An
application of concern would be if a user would generate
offensive caricatures and cartoons of other people without
consent. This can be used to insinuate biases against people or can have a detrimental effect on a person’s autonomy,
dignity, and/or privacy. The images used in this work are
taken or derived from the FFHQ dataset which has appropriate licenses for non-commercial research use and to
the best of our knowledge, is committed to protecting the
privacy of the individuals who do not wish to be included.