KYC and user verification: The most common reason for this to happen is inappropriate placement of your face within the frame for selfie. Make sure your face is in the centre of the frame and you are not too far from the camera. For the first selfie, you can look straight into the camera and for your second selfie, turn your face slightly to the left at an angle, not more than 45 degrees.
The liveness detection requires natural head movement to prevent “photo-swapping” attacks. Sudden or negligible head movements will cause a failure in liveness detection. To get your selfie verified, make sure your face is in the centre of the frame and the camera isn’t too far from the frame. Also, make natural head movements like nodding or slightly moving the face to the left at a 45 degree angle.
Please make sure that the head movement for the second selfie wasn’t too strong or too negligible as that leads to rejection of the selfie. To get yourself verified, do a nodding movement instead to avoid sudden head movements.
What you need to do:
- Do not use an image of your face but a real face.
- The user must maintain the appropriate distance and keep his or her face in the center of the image.
- The user must direct his or her nose to the desired location;
- The user must simulate a three-dimensional system by rotating his head.
How liveness detection works:
- It takes images of your face with a normal camera and examines them for modification and natural movement. The advanced algorithms (motion-analysis) detect the difference when you move inappropriately.
- Artificial intelligence is used to identify presentation attacks such as projections, 3D masks, and so on by leveraging powerful DCNNs (deep convolutional neural networks).
- It uses a particular texture-established algorithm to detect deep fakes. It detects when a resumed variant of a person is shown instead of the real thing.
- Optionally, it can direct the user to move their head in a random direction and confirm that the head was rotated in the correct direction.
