Vision Intelligence
Computer Vision is a technology for building an artificial system that obtains information from images or multi-dimensional data. The system is further equipped with intelligence by machine learning which involves algorithms that learn the pattern from data, to perform inference on new data. Machine learning comprises supervised, unsupervised, semi-supervised and reinforcement learning. One of the most important applications of computer vision is face detection and face recognition. We will see how supervised learning helps face recognition systems extract facial features and predict potential matches of faces.
Facial Recognition
A face recognition system takes an image of a face and makes a prediction about whether the face matches other faces in a provided database. Face recognition systems are designed to compare and predict potential matches of faces regardless of their expression, facial hair, and age. A wide variety of face recognition systems are available in the market. The one among fairly accurate and controversial facial recognition engines that works on many kinds of images and videos is called Amazon Rekognition.
Rekognition
Amazon Rekognition provides two distinct face capabilities using a type of machine learning called computer vision. The first capability is facial analysis for a particular image or video, the service can tell you where a face appears, and certain characteristics of the image. The second capability of Amazon Rekognition is commonly known as facial recognition. It is a distinct and different feature from facial analysis and attempts to match faces that appear similar.
In this tutorial, you will use Amazon Rekognition to analyze an image and then compare it to other images to see if the faces are the same. This tutorial is a demo of the functionality that is available when using the AWS CLI or the Rekognition API. For production or proof of concept implementations, please use programmatic interfaces rather than the Amazon Rekognition Console.
Step 1:
Open AWS management console by using either one of the following links. Create an account and post-login type Rekognition in the search bar and select Rekognition to open the service console.
us-east-2.console.aws.amazon.com/console/ho.. console.aws.amazon.com/console/home
Step 2:
In the left panel navigation, Object and scene detection, Image moderation, Facial analysis, Celebrity recognition, Face comparison, Text in image features are available to the playground. Select a respective feature to analyze and compare faces.
Analysis of facial attributes, including confidence scores
Matches based on a similarity percentage
Celebrities in images and provides confidence scores
Image Source: us-east-2.console.aws.amazon.com/rekognition
Using Javascript
The following JavaScript allows a user to choose an image and view the feature of faces that are detected in the image.
Please follow the below link for javascript code and steps involving identity pool creation that enables access to unauthenticated identities, and how to get the identity pool identifier that's needed in the example code. docs.aws.amazon.com/rekognition/latest/dg/r..
Credentials provider to use your identity pool
AWS.config.region = '{YOUR REGION}'; AWS.config.credentials = new AWS.CognitoIdentityCredentials({ IdentityPoolId: '{YOUR IDENTITY POOL}' }); // Make the call to obtain credentials AWS.config.credentials.get(function () { });
Call for Detect Faces API
var rekognition = new AWS.Rekognition();
var params = {
Image: {
Bytes: imageData
},
Attributes: [
'ALL',
]
};
rekognition.detectFaces(params, function (err, data) { console.log(data.FaceDetails);
});
Behind the scenes:
The biggest controversies involved in rekognition are inaccuracy in the analysis and selling of products to law enforcement. AI researchers from Google, Facebook, Microsoft, and a number of top universities have called on Amazon to stop selling its facial recognition technology to law enforcement. According to Amazon, to date (over two years after releasing the service), they have had no reported law enforcement misuses of Amazon Rekognition.
The answer to anxieties over new technology is not to run ‘tests’ inconsistent with how the service is designed to be used and to amplify the test’s false and misleading conclusions through the news media. We are eager to continue to work with researchers, academics, and customers, to continuously improve as we evolve this important technology.
-Dr. Matt Wood, general manager of artificial intelligence at AWS
Developer Guide
The complete API reference for developers are available at aws.amazon.com/rekognition/resources
References
aws.amazon.com/rekognition aws.amazon.com/rekognition/resources docs.aws.amazon.com/rekognition/latest/dg/r.. aws.amazon.com/blogs/machine-learning/thoug..