伙计们 有没有视觉智能平台的人脸对比1：1的demo啊？

一、人脸比对算法-公有云API 1、接入即用能力 1）人脸1比1 https://vision.aliyun.com/experience/detail?spm=a2cvz.27725926.J_7524944390.29.59d150b5AGw2rT&tagName=facebody&children=CompareFace 2）人脸搜索1比N（自建比对库场景） https://vision.aliyun.com/experience/detail?spm=a2cvz.27725934.J_7524944390.31.157e50b5XNQlBw&tagName=facebody&children=SearchFace 2、比对开发原子能力 1）人脸检测和五官定位 https://vision.aliyun.com/experience/detail?spm=a2cvz.27717767.J_7524944390.8.66cd28d0BpsfbD&tagName=facebody&children=DetectFace 2）静态人脸活体检测 https://vision.aliyun.com/experience/detail?spm=a2cvz.27725922.J_7524944390.13.199850b5SEuIJ3&tagName=facebody&children=DetectLivingFace 二、人脸比对算法-离线SDK 1、人脸1比1、1比N离线SDK https://vision.aliyun.com/mobilesdk/offlinefaceserver?spm=a2cvz.27846226.J_7524944390.21.51d24f943kFyFm 2、人脸关键点离线SDK https://vision.aliyun.com/mobilesdk/face_landmark?spm=a2cvz.27726939.J_7524944390.12.6caa7aa5rJ0Lf1 可以看下线上api可以产品体验测试或者集成接入测试。，此回答整理自钉群“阿里云视觉智能开放平台咨询1群”

阿里云视觉AI人脸人体类目下的人脸比对1:1能力推荐使用SDK调用，支持多种编程语言，调用时请选择AI类目为人脸人体（facebody）的SDK包，文件参数通过SDK调用可支持本地文件及任意URL。

请求示例：

http(s)://facebody.cn-shanghai.aliyuncs.com/?Action=CompareFace      //更多关于访问域名（Endpoint）信息，请参见：https://help.aliyun.com/document_detail/143103.html
&QualityScoreThreshold=98.5
&ImageURLA=http://viapi-test.oss-cn-shanghai.aliyuncs.com/viapi-3.0domepic/facebody/CompareFace/CompareFace-right1.png
&ImageURLB=http://viapi-test.oss-cn-shanghai.aliyuncs.com/viapi-3.0domepic/facebody/CompareFace/CompareFace-left1.png
&ImageDataA=/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgK****
&ImageDataB=/9j/4AAQSkZJRgABAQAAAQABAAD/2wBDAAgGBgcGBQgHBwcJCQgQ****
&公共请求参数

JSON格式返回格式：

HTTP/1.1 200 OK
Content-Type:application/json

{
  "RequestId" : "B352CD68-C847-5C48-BE68-3F266356FDD7",
  "Data" : {
    "Thresholds" : [ 61, 69, 75 ],
    "RectBList" : [ 169, 48, 125, 162 ],
    "Confidence" : 88.4831771850586,
    "RectAList" : [ 157, 82, 147, 188 ],
    "QualityScoreA" : 99.65772247314453,
    "QualityScoreB" : 98.01177978515625,
    "IsMaskA" : 0,
    "IsMaskB" : 0,
    "LandmarksAList" : [ 201, 146, 266, 159, 222, 190, 187, 208, 255, 219 ],
    "LandmarksBList" : [ 199, 115, 248, 109, 215, 146, 206, 167, 260, 162 ]
  }
}