Preserving privacy by de-identifying face images

By Elaine M. Newton, Latanya Sweeney, Bradley A. Malin. 2005

602 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

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 Signal Processing

Citations in the corpus

Defeating Image Obfuscation with Deep Learning (2016, methodology): “The authors suggested a new de-identiﬁcation technique based on Newton et al. [31].”
I Know That Person: Generative Full Body and Face De-identification of People in Images (2017, background): “[21], who propose the k-same face de-identification algorithm.”
Fawkes: Protecting Privacy against Unauthorized Deep Learning Models (2020, background): “er line of work seeks to edit facial images so that human-like characteristics are preserved but facial recognition model accuracy is signiﬁcantly reduced. Methods used includek-means facial averaging[35], facial inpainting[51], andGAN-basedface editing[27,52,64]. Since these dramatically alter the ”
Model-Based Face De-Identification (2006, methodology): “Unlike the methods discussed above, [21] also addresses the problem of adaptive recognition in which an adversery mimics the obfuscation technique employed by a privacy-protection algorithm.”
On the (In)effectiveness of Mosaicing and Blurring as Tools for Document Redaction (2016, background): “Newton, Sweeney, and Malin showed that face recognition software can be used to recognize mosaiced faces [15] from still images.”
Semi-adversarial Networks: Convolutional Autoencoders for Imparting Privacy to Face Images (2017, background): “On one hand, there are face de-identification techniques [11, 20, 7] where a face image is modified in order to confound a face matcher.”
A Hybrid Deep Learning Architecture for Privacy-Preserving Mobile Analytics (2017, background): “A fundamental work in this category is presented in [33], which targets privacy issue in video surveillance data.”
A General Survey of Privacy-Preserving Data Mining Models and Algorithms (2008, background): “More detailed discussions on some of these sssues may be found in [96, 114–116].”
Deep Private-Feature Extraction (2018, background): “Newton et al. [20] published a k - anonymous image dataset by proposing the k -same algo-rithm.”
Efficient privacy preserving video surveillance (2009, background): “Face swapping [5] and face deidentification [10, 13] try to modify face images such that they can be automatically detected, but cannot be correctly recognized.”
Efficient Privacy-Preserving Face Recognition (2009, neutral): “In order to preserve privacy, faces can be de-identified such that face recognition software cannot reliably recognize de-identified faces, even though many facial details are preserved as described in [28].”
Live Face De-Identification in Video (2019, neutral): “Earlier work implemented different types of image distortions for face deidentification [33, 10], while more recent works rely on techniques for selecting distant faces [41] or averaging/fusing faces from pre-existing datasets [32, 16, 31].”
A Scanner Darkly: Protecting User Privacy from Perceptual Applications (2013, methodology): “published several papers [11, 12, 19] on “de-identifying” datasets of face images.”