AWS Rekognition
1 min read
aws
Written in 2019. Tools, APIs and prices may have changed since.
Lab note What I wrote down while learning this by building with it. Short by design; the code is in pfeilbr/aws-rekognition-playground.
learn AWS Rekognition
see src/examples.js
Prerequisites
AWS default credentials available to call Rekognition API
install imagemagick, graphicsmagick, and ghostscript (for fonts) for labelling image
brew install imagemagick brew install graphicsmagick brew install ghostscript
Example Labelled Image Using detectLabels API
Source
Labelled
Example Labelled Image Using detectText API
Source
Labelled
Running
npm install
npm start
Video Label Detection Example via CLI
# start (it's async)
aws rekognition start-label-detection --video "S3Object={Bucket=rekognition-playground,Name=SampleVideo_1280x720_1mb.mp4}"
# output
{
"JobId": "2c9b387607977af21c0839f177bf7034ce1bcd5139810b533dea3deb6361f348"
}
# in progress
aws rekognition get-label-detection --job-id "2c9b387607977af21c0839f177bf7034ce1bcd5139810b533dea3deb6361f348"
# output
{
"Labels": [],
"LabelModelVersion": "2.0",
"JobStatus": "IN_PROGRESS"
}
# completed
aws rekognition get-label-detection --job-id "2c9b387607977af21c0839f177bf7034ce1bcd5139810b533dea3deb6361f348"
# output
{
"Labels": [
{
"Timestamp": 0,
"Label": {
"Instances": [],
"Confidence": 89.929931640625,
"Parents": [],
"Name": "Animal"
}
},
{
"Timestamp": 0,
"Label": {
"Instances": [],
"Confidence": 69.0970458984375,
"Parents": [
{
… 3753 more lines — see the full README.