Using Rekognition in Business Coaching video analysis


Hi, I am a business coach and I use AI to support my coaching programmes. I am trying to discover whether I could use Rekognition facial recognition to analyses the video of a client in a coaching session to get timestamps for visible reactions (smiling / frowning/ long pauses etc.) I am new to the topic and I chatted with Kenneth in your support chat and he advised that I create a free tier account (done) and raise a case with a Cloud Support Engineer to confirm whether video is included in the Free Tier Trial and, if it is, how I can learn to use Rekognition to detect facial cues in my recorded videos. He also advised I resend this question via AWS re:Post. Thanks for your help!

1 Answer

Hi Eruditic,

Please go through below steps and documentation link it will help to resolve your issue.

Confirming Free Tier Availability

To confirm whether video analysis is included in the Free Tier, follow these steps:

Create a Support Case: Log in to your AWS Management Console. Navigate to the Support Center. Click on "Create Case". Select "Account and Billing Support". In the case description, mention that you want to confirm if video analysis with Amazon Rekognition is included in the Free Tier and how it can be utilized.

Understanding Amazon Rekognition Video Analysis Amazon Rekognition provides a range of capabilities for video analysis, including the detection of faces and facial attributes (e.g., emotions like smiling or frowning).

Setting Up Rekognition for Video Analysis Here are the steps to analyze videos using Amazon Rekognition:

Upload Your Video to Amazon S3:

  • Ensure your video is in a supported format (e.g., MP4).
  • Upload the video to an S3 bucket in your AWS account.

Initiate Video Analysis:

  • Use the Rekognition API to start a video analysis job. This can be done using the AWS SDK for Python (Boto3) or directly through the AWS Management Console.

Here is an example using Python:

import boto3

# Create a Rekognition client
rekognition_client = boto3.client('rekognition')

# Start label detection
response = rekognition_client.start_face_detection(
    Video={'S3Object': {'Bucket': 'your-bucket-name', 'Name': 'your-video-file.mp4'}}

# Get the job ID
job_id = response['JobId']
print(f'Started face detection job with ID: {job_id}')

Check the Status of the Job: You need to periodically check the status of your job until it's complete.

import time

while True:
    response = rekognition_client.get_face_detection(JobId=job_id)
    status = response['JobStatus']
    if status in ['SUCCEEDED', 'FAILED']:
    print(f'Job status: {status}')
    time.sleep(10)  # Wait before checking again

if status == 'SUCCEEDED':
    print('Face detection completed successfully')
    print('Face detection failed')

Retrieve and Analyze Results: Once the job is completed, you can retrieve the results.

response = rekognition_client.get_face_detection(JobId=job_id)

for face in response['Faces']:
    timestamp = face['Timestamp']
    emotions = face['Face']['Emotions']
    for emotion in emotions:
        print(f"Timestamp: {timestamp} ms, Emotion: {emotion['Type']}, Confidence: {emotion['Confidence']:.2f}%")

Interpreting the Results

  • Timestamp: The time in milliseconds from the start of the video when the face was detected. *** Emotions:** List of detected emotions with their confidence level

AWS Documentation:

answered a month ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions