Differ between real human face and static photo in iOS swift? [closed]

Clash Royale CLAN TAG#URR8PPP
Differ between real human face and static photo in iOS swift? [closed]
How can I use camera to differentiate between real human face and static photo using iPhone X TrueDepth camera in Swift? Any sample project to do that? Any way to implement that please advise.
Thank you!
This question appears to be off-topic. The users who voted to close gave this specific reason:
You could try ARKit
– Paulw11
Aug 10 at 8:52
@Paulw11 Tried it but its also detecting face from a picture.
– starptr
Aug 10 at 9:18
@SameeShah If you are able to collect enough depth samples of real faces vs photos, then you could train a CoreML model that would predict that. Considering that face depth and photo depth (which is a surface basically) are quite different, it should work fairly well. But that's far from an out-of-the-box solution of course.
– Lësha Turkowski
Aug 11 at 10:43
You are right but I dont know how do I get face depth using iPhone camera. The support for 3D scan is coming on iOS 12 beta with ARKit. Any suggestions on that?
– starptr
Aug 12 at 11:03
1 Answer
1
Take a look at Apple's Vision Framework
. Here is a sample project , not exactly what you need but you can do some tweaking if needed
EDIT:
try this one. If you want to differentiate between human face and static photo you have to use blink detection or something similar
Its not what I am looking for. It can't differentiate between static photo and live human. It detects faces only.
– starptr
Aug 10 at 9:18
edited my answer
– Linh Ta
Aug 10 at 10:29
Already tried that. Don't differentiate if I record a video and show it to camera. It detects blink from real human face as well as from video.
– starptr
Aug 10 at 10:45
I don't think there is any API for TrueDepth camera or access to the hardware that provides this
– Scriptable
Aug 10 at 8:27