Overview
\nWelcome to the FaceX API Documentation v1.0.
If you’re new to FaceX, create a free account and follow this guide to get started.
\nThe FaceX API is built on REST principles. Authenticated users can interact with any of our URIs by using the specified HTTP request method. We recommend and also enforces our users to use SSL encryption by issuing requests through HTTPS. Our global endpoint for all APIs:
\nhttp://106.51.58.118:5000
\nFaceX do provide to help it’s customer meeting the local legal and compliance.
\nPrerequisites
\nWhile we have PHP, NodeJS, Python, Java, C#, Perl and Ruby Sample Codes for our APIs, but it is always good to test the API before going into integration with your application. For testing, we recommend using cURL.
Note:Curl comes standard on Mac operating systems. On other OS, you might have to install the same.
\nBefore you can start using the API, you need to do the following:
\nWith FaceX API's you can now add Face Recognition capabilities to your App using our simple API Calls. Our services are implemented through state of the art Deep learning techniques trained over millions of images to offer you the Highest Accuracy with the quickest turn-around time. Our services are perfect fit for you if you plan to have Face ID, Gender and Age Detection or even for face transformations (Masking/Swapping etc).
\nAll our services are currently hosted under our own secure cloud, processing with the latest rack of GeForce GPUs, promising you a result within 0.02sec and an up-time of 99.99%.
\nThe three APIs that we currently support are:
\nget_image_attr : For detecting faces , gender detection , age range prediction and facial landmark
\ncompare_faces : For comaparing two faces to check if it belongs to the same person
\nget_face_vec : Get 128 dimensional feature vector describing a face for recognition and identification purpose.
\nUser Id
\n"},{"key":"user_key","value":"98e739ec2c949fd43085","description":"User Password
\n"}],"body":{"mode":"formdata","formdata":[{"key":"image_attr","value":"","description":"multipart/base64 encoded image or remote url of image
\n","type":"text"}]},"url":"http://106.51.58.118:5000/get_image_attr?face_det=0&bboxes=120,150,200,250","description":"Get Image Attributes
\nUse facial attributes api to detect faces, facial landmark, gender and age.
\nIn example request code generated by postman the payload should be changed as:\n
payload = {\"image_attr\":\"multipart/base64 encoded image or remote url of image\"}\nWe recommend an image resolution no higher than 1024x, since it could slow down the request process. Supported image formats are JPEG/PNG format.
\nFor enabling face detection set value as 1 else 0
\n","type":"text/plain"},"key":"face_det","value":"0"},{"description":{"content":"This parameter is optional.
If bounding box coordinates are available set 'face_det' to '0' and set values for key 'bboxes' in format x1,y1,x2,y2 .\n'bboxes' values should be passed as string seperated by '&' \nex:\"120,150,200,250&130,180,240,300\" or for just one face pass as \"120,150,200,250\"
User Id
\n"},{"key":"user_key","value":"98e739ec2c949fd43085","description":"User Password
\n"}],"body":{"mode":"formdata","formdata":[{"key":"img_1","value":null,"description":"multipart/base64 encoded image or remote url of image
\n","type":"file"},{"key":"img_2","value":null,"description":"multipart/base64 encoded image or remote url of image
\n","type":"file"}]},"url":"http://106.51.58.118:5000/compare_faces?face_det=0&bboxes=120,150,200,250","description":"Compare if two faces belong to same person and gives confidence score
\nReturns a confidence value which is greater than 0.4 then it is a match
\nFor better results,face standard such as 90 pixels from eye to eye is required.
\nIn example request code generated by postman the payload should be changed as:
\npayload = {\"img_1\":\"multipart/base64 encoded image or remote url of image\",\"img_2\":\"multipart/base64 encoded image or remote url of image\"}\nWe recommend an image resolution no higher than 1024x1024, as it could slow down the request process. Supported image formats are JPEG/PNG format.
\nFor enabling face detection set value as 1 else 0
\n","type":"text/plain"},"key":"face_det","value":"0"},{"description":{"content":"This parameter is optional.
If bounding box coordinates are available set 'face_det' to '0' and set values for key 'bboxes' in format x1,y1,x2,y2 .\n'bboxes' values should be passed as string seperated by '&' \nex:\"120,150,200,250&130,180,240,300\" or for just one face pass as \"120,150,200,250\"
Uer Id
\n"},{"key":"user_key","value":"98e739ec2c949fd43085","description":"User Password
\n"}],"body":{"mode":"formdata","formdata":[{"key":"img","value":null,"description":"multipart/base64 encoded image or remote url of image
\n","type":"file"}]},"url":"http://106.51.58.118:5000/get_face_vec?face_det=0&bboxes=130,180,240,300","description":"Get 128-D feature vector extracted from the face.
\nUpload a image containing face
\nFeature vector of one person can be compared with feature vector of another person using euclidean distance:If F1 and F2 are feature vector corresponding to two faces then:
\n if euclidean(F1,F2)>Threshold(=0.6):\n Same Person\n \nExample:
\nFeature vector = [-0.133286, 0.0671902, 0.0648706, ............., -0.0325235, -0.0260845, 0.0237684, -0.0272609, -0.106857]
\npayload = {\"img\":\"multipart/base64 encoded image or remote url of image\"}For enabling face detection set value as 1 else 0
\n","type":"text/plain"},"key":"face_det","value":"0"},{"description":{"content":"This parameter is optional.
If bounding box coordinates are available set 'face_det' to '0' and set values for key 'bboxes' in format x1,y1,x2,y2 .\n'bboxes' values should be passed as string seperated by '&' \nex:\"120,150,200,250&130,180,240,300\" or for just one face pass as \"120,150,200,250\"