Face Recognition Vendor Test (FRVT) Part 4: MORPH - Performance of Automated Face Morph Detection

Published

March 6, 2020

Author(s)

Mei L. Ngan, Patrick J. Grother, Kayee K. Hanaoka, Jason M. Kuo

Abstract

The FRVT MORPH test provides ongoing independent testing of prototype face morphing attack detection (MAD) technologies. The evaluation is designed to obtain commonly measured assessment of morph detection capability to inform developers and current and prospective end-users. FRVT MORPH is open for ongoing participation worldwide, and there is no charge to participate. The test leverages a number of datasets created using different morphing methods with goals to evaluate algorithm performance over a large spectrum of morphing techniques. Testing was conducted using a tiered approach, where algorithms were evaluated on low quality morphs created with readily accessible tools available to non-experts, morphs generated using automated morphing methods based on academic research, and high quality morphs created using commercial- grade tools. We’d like to get an assessment on the existence and extent of morph detection capabilities, and if there is indication of high accuracy, much larger datasets can be curated to support large-scale evaluation of the technology.

Citation

NIST Interagency/Internal Report (NISTIR) - 8292

Report Number

8292

NIST Pub Series

NIST Interagency/Internal Report (NISTIR)

Pub Type

NIST Pubs

Download Paper

DOI Link

Keywords

face morphing, biometrics, face recognition

Citation

Ngan, M. , Grother, P. , Hanaoka, K. and Kuo, J. (2020), Face Recognition Vendor Test (FRVT) Part 4: MORPH - Performance of Automated Face Morph Detection, NIST Interagency/Internal Report (NISTIR), National Institute of Standards and Technology, Gaithersburg, MD, [online], https://doi.org/10.6028/NIST.IR.8292 (Accessed September 28, 2022)