AI handily beats humans at biometric spoof attack detection in ID R&D research

Biometric spoofing attacks are more easily spotted by artificial intelligence-based computer systems than by people, according to new research published by ID R&D.

The new report, ‘Human or Machine: AI Proves Best at Spotting Biometric Attacks,’ compares the relative effectiveness of humans and computers detecting presentation attacks, in terms of speed and accuracy. Liveness detection was tested against images including spoof attempts with printed photos, videos, digital images, and 2D or 3D masks, according to the announcement.

The company’s IDLive Face accepted 0 percent of face biometric spoofs across all types of attacks and 175,000 images. People fared far worse, failing to spot spoofs in every category, including 30 percent of photo prints, one of the easiest spoof attacks for fraudsters to carry out. When a group of 17 people were polled for collective decisions, the human accuracy rates improved, but did not beat the computer’s performance.

Humans also took longer to determine liveness; 4.8 seconds per image compared to less than 0.5 seconds for computers running a single CPU.

Humans participating in the test classified 18 percent genuine faces as spoofs, while the AI system incorrectly classified only 1 percent of genuine faces as biometric spoofs. ID R&D says that many facial liveness systems which are effective at detecting frauds also block a significant number of real customers.

ID R&D will present a webinar to explore how businesses can improve the accuracy of their biometric liveness detection to avoid this situation on February 22, 2022. The webinar will include a deep dive into the research, its motivations and methodology, a closer look at the spoof attacks used in the experiment and the data generated.

“The results are undeniable,” states Alexey Khitrov, CEO at ID R&D. “Biometric technology used for identity verification has evolved in recent years to increase speed and accuracy, now significantly outperforming the human eye. Organizations can achieve tremendous efficiencies by using identity verification systems that include a biometric component. However, there is still work to be done and we are excited to see biometrics helping to build consumer trust.”

Khitrov will also host a Q&A and preview further research being conducted by ID R&D during the webinar.

ID R&D has reported dramatic increases in the use of IDLive Face for passive liveness detection as selfie biometrics have been widely adopted for remote identity proofing.

Article Topics

accuracy  |  AI  |  biometric liveness detection  |  biometrics  |  face biometrics  |  ID R&D  |  identity verification  |  passive facial liveness  |  presentation attack detection  |  spoof detection  |  webinar