• Corpus ID: 195750834

A Utility-Preserving GAN for Face Obscuration

  title={A Utility-Preserving GAN for Face Obscuration},
  author={Hanxiang Hao and David G{\"u}era and Amy R. Reibman and Edward J. Delp},
From TV news to Google StreetView, face obscuration has been used for privacy protection. Due to recent advances in the field of deep learning, obscuration methods such as Gaussian blurring and pixelation are not guaranteed to conceal identity. In this paper, we propose a utility-preserving generative model, UP-GAN, that is able to provide an effective face obscuration, while preserving facial utility. By utility-preserving we mean preserving facial features that do not reveal identity, such as… 

Figures and Tables from this paper

StyleGAN as a Utility-Preserving Face De-identification Method

This paper investigates the use of StyleGAN in generating de-identified faces through style mixing, where the styles or features of the target face and an auxiliary face get mixed to generate a de-identified face that carries the utilities of thetarget face.

Learning Disentangled Representations for Identity Preserving Surveillance Face Camouflage

This paper introduces a novel individual face protection model, which can camouflage the face appearance from the perspective of human visual perception and preserve the identity features of faces used for face authentication and validate its superiority in terms of visual quality and scalability.

DeepBlur: A Simple and Effective Method for Natural Image Obfuscation

DeepBlur is presented, a simple yet effective method for image obfuscation by blurring in the latent space of an unconditionally pre-trained generative model that is able to synthesize photo-realistic facial images.

Robustness Analysis of Face Obscuration

This paper proposes to measure the effectiveness of eight obscuration techniques by attacking the redacted faces in three scenarios: obscured face identification, verification, and reconstruction and shows that the k-same based methods are the most effective.

Identity-Preserving Face Anonymization via Adaptively Facial Attributes Obfuscation

A face anonymization framework that could obfuscate visual appearance while preserving the identity discriminability and can achieve a balance between recognition utility and appearance anonymizing by modifying different numbers of facial attributes according to pratical demands.

Protecting image privacy through adversarial perturbation

The results show that the proposed method performs remarkably well on preventing privacy from being exposed by DNN detectors, while causing very limited degradation to the visual quality of images.

Privacy–Enhancing Face Biometrics: A Comprehensive Survey

The goal of this overview paper is to provide a comprehensive introduction into privacy–related research in the area of biometrics and review existing work on Biometric Privacy–Enhancing Techniques (B–PETs) applied to face biometricrics.

Robust Privacy-Preserving Motion Detection and Object Tracking in Encrypted Streaming Video

This paper analyzes the properties of the video codec and format-compliant encryption schemes to propose a new compressed-domain feature to capture motion information in complex surveillance scenarios and proposes an efficient and robust privacy-preserving motion detection and multiple object tracking scheme.

Determining the Robustness of Privacy Enhancing DeID Against the ReID Adversary: An Experimental Study

A novel approach is described, in which a selected ReID algorithm is pitted against multiple DeID techniques to test the robustness of these DeID methods, and to determine if ReID can pose a threat to DeID as an adversary.


This thesis considers statistical steganalysis of images in two different frameworks, first study staganalysis in the framework of statistical hypothesis testing, and proposes a hiding scheme using a reference matrix to lower the distortion caused by embedding.

Face aging with conditional generative adversarial networks

This work proposes the first GAN-based method for automatic face aging and introduces a novel approach for “Identity-Preserving” optimization of GAN's latent vectors.

Integrating Utility into Face De-identification

A new algorithm, k-Same-Select, is introduced, which is a formal privacy protection schema based on k-anonymity that provably protects privacy and preserves data utility and empirically validate the findings through evaluations on the FERET database.

Natural and Effective Obfuscation by Head Inpainting

This work proposes a novel head inpainting obfuscation technique that generates realistic person images, while achieving superior obfuscation performance against automatic person recognizers.

Fast Face-Swap Using Convolutional Neural Networks

A new loss function is devised that enables the network to produce highly photorealistic results by making face swap work in real-time with no input from the user.

k-Same-Net: k-Anonymity with Generative Deep Neural Networks for Face Deidentification †

This work proposes a novel approach towards face deidentification, called k-Same-Net, which combines recent Generative Neural Networks (GNNs) with the well-known k-Anonymitymechanism and provides formal guarantees regarding privacy protection on a closed set of identities.

GARP-face: Balancing privacy protection and utility preservation in face de-identification

This paper uses modern facial analysis technologies to determine the Gender, Age, and Race attributes of facial images, and preserves these attributes by seeking corresponding representatives constructed through a gallery dataset, and shows that the proposed approach outperforms previous solutions in preserving data utility while achieving similar degree of privacy protection.

Attribute-Guided Face Generation Using Conditional CycleGAN

This work condition the CycleGAN and proposes conditional CycleGAN, which is designed to handle unpaired training data because the training low/high-res and high-res attribute images may not necessarily align with each other, and to allow easy control of the appearance of the generated face via the input attributes.

Defeating Image Obfuscation with Deep Learning

It is shown how to train artificial neural networks to successfully identify faces and recognize objects and handwritten digits even if the images are protected using any of the above obfuscation techniques.

Preserving privacy by de-identifying face images

A new privacy-enabling algorithm is presented, named k-Same, that guarantees face recognition software cannot reliably recognize deidentified faces, even though many facial details are preserved.

Age Progression/Regression by Conditional Adversarial Autoencoder

A conditional adversarial autoencoder that learns a face manifold, traversing on which smooth age progression and regression can be realized simultaneously is proposed, and the appealing performance and flexibility of the proposed framework is demonstrated by comparing with the state-of-the-art and ground truth.