Deep Privacy Aims to Anonymize People While Retaining Expressions

I of the most important concerns in deepfakes is undoubtedly privacy, which is exactly what these researchers at the Norwegian University of Science and Technology has tried to fix.

This new technique named Deep Privacy makes employ of generative adversarial networks(GAN), the underlying technology employed in Deep fakes to anonymize the subjects while replicating their characteristics.

The algorithms used in Deep Privacy extract critical facial expressions from the subject and runs it through the database trained with 1.5 million face images to generate a new face retaining all the details from the source paradigm.

For accurately detecting facial features, the researchers utilise Mask R-CNN and Dual Shot Confront Detector (DSFD). Mask R-CNN is responsible for generating thin pose information of the face up while DSFD is used for detecting the faces nowadays in the image.

"Nosotros present experimental results reflecting the capability of our model to anonymize images while preserving the data distribution, making the information suitable for further training of deep learning models. As far as we know, no other solution has been proposed that guarantees the anonymization of faces while generating realistic images.", wrote the researchers.

Since the project is highly experimental at this point in time, y'all should not exist expecting country-of-the-art results here. From the implementation, even so, it appears like information technology could improve significantly over time.

The source code of the research paper is available on GitHub. If y'all're interested to run the code yourself, you may practice so from here. The instructions to set up the environment are given in the README.md file in the repository as well. You can besides run the project in Google Colab from here.

So, what do yous think of Deep Privacy? Let u.s. know your thoughts in the comments.