AWS-Rekognition for Video Analytics Using Python
We here implement Advance Scene Detection Analytics across Edge and Cloud resources.The proposal uses AWS(Amazon Web Services) as a base platform for implementation. It is an attempt to mimic the scenario described in the paper Demonstration of a Cloud-based Software Framework for Video Analytics Application using Low-Cost IoT Devices. For more detailed explantation of paper watch this video A Cloud-based Smart Doorbell using Low-Cost COTS Devices.
The Salient Features of implementaion are :
- Known/Unknown Face Detection
- Animal Detection like Dog, Cat, Kangaroo etc.
- Unsafe Content Detection like Knife, Guns, Weapons.
- NoteWorthy Vehicle Detections like Ambulance, Fire Truck, Courier Vans (FedEx, DHL etc.)
AWS Services Used
- AWS Rekognition - For advance scence detection in a video
- AWS Kinesis - For uploading video analytics data of edge to AWS cloud.
- AWS DynamoDB - For storing video analytics data in Cloud.
- AWS S3- For storing videos and frames (images) of edge at Cloud.
- AWS Lambda - For handling all events of Cloud.
Configuration of AWS Services
1. AWS Kinesis : Create a data stream to get data from edge.
2. AWS DynamoDb :
Create a Table in DynamoDB with partion key as "frame_id".
Create Table with following cloumns :
3. AWS S3 :
Create a S3 bucket and create two folders with names “frames” & “video” in it.
4. AWS Lambda : Create a Lambda Function with all permissions IAM Role and add “Kinesis” & “DynamoDB” as Triggers. Paste the lamda_function.py code in your lambda function
Edge Implementation
Run main-video-analytic-code.py on Edge Device(Laptop, Raspberry PI)
Output
- Edge Output
- DynamoDB Output
