Company D has a legacy employee verification product which exists for many years. The system has used some old facial recognition technologies by comparing new photos with original ones in a large file system to verify employees. It is becoming harder and harder to maintain the system as all product architects have left and technical documents are not well maintained. The company decides to migrate the system to AWS and use some new technologies if possible. However, they do not want to spend huge efforts on the migration and hope to finish it as soon as possible. Which option is the best for the company to choose at a reasonable cost?

Answer options:

A.Utilize a popular facial recognition service that has used new biometric matching technology. Put the application in EC2 to handle the facial recognition tasks. Also, migrate the on-premise file system to S3.
B.Use Rekognition CLI to develop an application to use AWS Rekognition facial analysis service. For example, “aws rekognition detect-faces” can return a JSON response that contains information for each detected face. It can also compare faces to see if they match.
C.Utilize a modern facial recognition SAAS. Put the application in EC2 or lambda to communicate with the SAAS service provider. Also, migrate the on-premise file system to RDS. Make sure that the SAAS provider has access to RDS.
D.Transform all source or target pictures to base64-encoded bytes. Use Rekognition CLI to compare faces such as “aws rekognition compare-faces” by using the pictures with the format of image bytes.

Correct Answer – B
 
AWS always recommends exploring new services that can help increase working efficiency. Amazon Rekognition is a service that makes it easy to add image analysis to your applications. With Rekognition, you can detect objects, scenes, and faces in images. You can also search and compare faces. Rekognition’s API lets you easily build powerful visual search and discovery into your applications. With Amazon Rekognition, you only pay for the images you analyze and the face metadata you store. There are no minimum fees, and there are no upfront commitments. As the question asks for a migration with a reasonable cost, it is not proper to use the SAAS provider.
The below is a piece of an example response of a DetectFaces API call. Lots of information can be returned by Rekognition:
{
"FaceDetails": [
{
"AgeRange": {
"High": 43,
"Low": 26
},
"Beard": {
"Confidence": 97.48941802978516,
"Value": true
},
"BoundingBox": {
"Height": 0.6968063116073608,
"Left": 0.26937249302864075,
"Top": 0.11424895375967026,
"Width": 0.42325547337532043
},
"Confidence": 99.99995422363281,
 "Emotions": [
{
"Confidence": 0.042965151369571686,
"Type": "DISGUSTED"
},
{
"Confidence": 0.002022328320890665,
"Type": "HAPPY"
},
{
"Confidence": 0.4482877850532532,
"Type": "SURPRISED"
},
…
}
Option A is incorrect: Because the price will be higher than Rekognition. It is not the best choice.
Option B is CORRECT: Because Rekognition can definitely meet the need with a reasonable cost. About the facial recognition function of Rekognition, please refer to https://aws.amazon.com/getting-started/tutorials/detect-analyze-compare-faces-rekognition/ The compare-faces CLI reference is in https://docs.aws.amazon.com/cli/latest/reference/rekognition/compare-faces.html
Option C is incorrect: Same reason as A. Also, RDS is unsuitable for storing photos.
Option D is incorrect: It is quite similar to option
B. However, for Rekognition CLI, compare-faces only support source and target pictures in S3. Refer to the below in https://docs.aws.amazon.com/cli/latest/reference/rekognition/compare-faces.html:
“If you use the AWS CLI to call Amazon Rekognition operations, passing image bytes isn`t supported. The image must be formatted as a PNG or JPEG file.”