Question: Which one of these next 3 pictures is genuine?
Response: None. They are all spoofs.
The original images are from wikimedia and can be found here#1, here#2 and here#3. We recaptured the original images using a Galaxy S10+ with 4K resolution on an iMac with 4K retina display. Then, we cropped the images. The entire images can be found here#4, here#5 and here#6.
These kind of high resolution (quality) Print-Attacks are hard to detect and most liveness detectors would fail the test. Our passive (frictionless) face liveness detector uses SOTA (State Of The Art) deep learning techniques to spot both Print-Attack and Replay-Attack. You can freely test our implementation with your own images at https://www.doubango.org/webapps/face-liveness/
We’re working to package and release the source code of the SDK. This document will be updated to include the API reference and a Getting Started guide.
Next video shows the liveness detector on Replay-Attack:
You can also check our online cloud-based implementation (no registration required) at https://www.doubango.org/webapps/face-liveness/ to check out the accuracy and precision before starting to play with the SDK.
Cloud-based implementation: https://www.doubango.org/webapps/face-liveness/
Open source Computer Vision library: https://github.com/DoubangoTelecom/compv
- Supported attacks
- Why liveness detector is required for facial recognition systems?
- Passive versus Active liveness detection
- Improving the recall score
- Integration with your existing application
- Error and success codes
- Known issues