A Hybrid Model for Identity Obfuscation by Face Replacement

By Qianru Sun, Ayush Kumar Tewari, Weipeng Xu, Mario Fritz, Christian Theobalt, Bernt Schiele. 2018

135 citations recorded by the source corpus.

As more and more personal photos are shared and tagged in social media, avoiding privacy risks such as unintended recognition, becomes increasingly challenging. We propose a new hybrid approach to obfuscate identities in photos by head replacement. Our approach combines state of the art parametric face synthesis with latest advances in Generative Adversarial Networks (GAN) for data-driven image synthesis. On the one hand, the parametric part of our method gives us control over the facial parameters and allows for explicit manipulation of the identity. On the other hand, the data-driven aspects allow for adding fine details and overall realism as well as seamless blending into the scene context. In our experiments we show highly realistic output of our system that improves over the previous state of the art in obfuscation rate while preserving a higher similarity to the original image content.

Subfields: Computer Vision and Pattern Recognition, Privacy & Security

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Privacy & Security

Citations in the corpus

Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI (2019, neutral): “These kind of explanations have motivated, for instance, secure matching cryptographic protocols for photographer and bystanders to preserve privacy [356,421,422] .”
Live Face De-Identification in Video (2019, unknown): “In the GAN-based methods of [43, 44], face de-identification is employed for the related task of person obfuscation.”
Face identity disentanglement via latent space mapping (2020, methodology): “Tewari et al. [2020] utilize a pre-trained 3DMM to learn semantic face edits in latent space.”
Disentangling in Latent Space by Harnessing a Pretrained Generator (2020, methodology): “remove the face from the target image and complete it using a GAN [54], while other methods [19, 61] shift the identity using a pre-trained face recognition network, while keeping the general image relatively similar to the source.”
The Creation and Detection of Deepfakes (2020, background): “Instead of swapping identities, the authors of [151] wanted to variably obfuscate the target’s identity.”
CIAGAN: Conditional Identity Anonymization Generative Adversarial Networks (2020, methodology): “The results of [35] are visually appealing, but due to an optimization procedure for face alignment, the method is not”
Fawkes: Protecting Privacy against Unauthorized Deep Learning Models (2020, methodology): “Some have proposed distorting images to make them unrecognizable and thus avoiding facial recognition [26, 51, 63].”