Google Vision Api:区分真实面孔和照片 [英] Google Vision Api: Differentiate Real Face and A Photo
问题描述
我正在尝试开发一个面部识别系统.但是问题是它可以被照片绕过.我正在使用Google视觉api来检测人脸.有没有一种方法可以避免检测照片中的人脸?只想知道镜头前是否有一个真实的人.
I'm trying develop a facial recognition system. But the issue is it can be by-passed by a photo. I'm using google vision api to detect faces. Is there a way to avoid detecting faces in a photo? Just want to know if there is a real person standing in front of the camera.
推荐答案
Google Vision API不支持区分真实面孔和照片中的面孔.
The Google Vision API does not support distinguishing between a real face and a face in a photo.
通常,您可以尝试多种方法来执行此操作.也许最可靠的方法是使用立体摄像机或3D深度传感器从不同的角度推断面部的3D结构-真实的面部和面部照片显然具有不同的3D结构.
There are a number of ways that you might try to do this in general. Perhaps the most reliable way would be to use stereo cameras or 3D depth sensors to infer the 3D structure of the face from different viewpoints -- a real face and a photo of a face would obviously have a different 3D structure.
这篇关于Google Vision Api:区分真实面孔和照片的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!