Surface nanostructures for engineering spectral responsivity in imaging sensor pixels for combatting deepfakes

Lisa McPhillips; Ahasan Ahamed; Cesar Bartolo-Perez; M. Saif Islam

doi:10.1117/12.2597841

1 August 2021 Surface nanostructures for engineering spectral responsivity in imaging sensor pixels for combatting deepfakes

Lisa McPhillips, Ahasan Ahamed, Cesar Bartolo-Perez, M. Saif Islam

Author Affiliations +

Proceedings Volume 11841, Optics and Photonics for Information Processing XV; 118410G (2021) https://doi.org/10.1117/12.2597841
Event: SPIE Optical Engineering + Applications, 2021, San Diego, California, United States

Abstract

Current deepfake production methods use auto-encoders augmented by a generative adversarial network (GAN) to create fraudulent but convincing video footage. Developing neural networks to counteract these deepfakes is a highly active area of research—but software-based methods can be immediately used to benchmark even better deepfakes. Thus, there is a need for hardware based solutions to complement existing deepfake detection methods. Here, we present on-chip silicon spectrometer arrays to enhance the number of color channels detected in the imaging system by a factor of 100. These arrays are made up of unique photodiodes engineered to have distinctive spectral responsivities that arise from their photon-trapping, surface based, nanostructures. Videos recorded with this hyperspectral imaging device could complicate the training process for deepfake producers because it collects information that a standard camera cannot. It could also assist novel authentication methods, such as heartbeat monitoring, camera fingerprinting techniques, etc. These spectrometer arrays show a promising direction for continued research in deepfake detection.

Citation Download Citation

Lisa McPhillips, Ahasan Ahamed, Cesar Bartolo-Perez, and M. Saif Islam "Surface nanostructures for engineering spectral responsivity in imaging sensor pixels for combatting deepfakes", Proc. SPIE 11841, Optics and Photonics for Information Processing XV, 118410G (1 August 2021); https://doi.org/10.1117/12.2597841

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available