Adversarial Face
Adversarial face research focuses on creating subtly altered images that fool facial recognition systems, primarily to protect privacy or evaluate system robustness. Current research explores generating these "adversarial faces" using generative adversarial networks (GANs), diffusion models, and other techniques, often focusing on improving the imperceptibility of the alterations and the transferability of attacks across different recognition models. This field is significant because it highlights vulnerabilities in facial recognition technology, impacting both the development of more robust systems and the ongoing debate surrounding privacy implications of widespread facial recognition deployment.
Papers
July 18, 2024
May 27, 2024
February 26, 2024
December 18, 2023
November 29, 2023
September 4, 2023
September 3, 2023
June 22, 2023
May 8, 2023
April 22, 2023
January 28, 2023
October 28, 2022
May 28, 2022
April 18, 2022
March 7, 2022