- objects (e.g. flower, tree, or table)
- events (e.g. a wedding, graduation, or birthday party)
- concepts (e.g. a landscape, evening, and nature)
The general prerequisites to use Amazon Rekognition are documented in Getting Started with Amazon Rekognition. The configuration is simpler that Amazon Comprehend as you don't need to use asynchronous operations.
See the list of supported AWS regions where Amazon Rekognition is available.
- Amazon Rekognition supports the PNG and JPEG image formats. So the images provided as input to various API operations, such as DetectLabels, must be in one of the supported formats.
- Images up to 5 MB are passed directly as raw bytes. Images over 5 MB go via an S3 bucket setup for Intelligence Services and Rekognition. The maximum image size allowed by Rekognition is limited to 15 MB.
- The minimum image size is 80 pixels for both height and width.
You'll need to create an AWS Identity and Access Management (IAM) role with the correct permissions to control access to AWS services and resources.
There is a setting for the level of confidence that Amazon Rekognition has in the accuracy of the extracted content. This is defined as the minimum confidence level and has a default value of 80%.
See Setting up services in AWS for more.