Thanks for your interesting work.
We are curious about the model's ability to reconstruct face embedding under the 0% CPD setting. This may be a good benchmark for measuring the security in face privacy method. Would you provide the pre-trained model of 0% CPD?
Thanks.
Thanks for your interesting work. We are curious about the model's ability to reconstruct face embedding under the 0% CPD setting. This may be a good benchmark for measuring the security in face privacy method. Would you provide the pre-trained model of 0% CPD? Thanks.