AWS Panorama for Computer Vision Applications
By: Viraj Deshwal, Machine Learning Engineer, TensorIoT and Nicholas Burden, Technical Evangelist, TensorIoT
AWS Panorama - AWS Panorama provides integration with the cloud to bring Computer Vision to the edge with a fully managed AWS Edge inference device for real-time inference.
The Amazon Web Services (AWS) re:Invent release of AWS Panorama has greatly facilitated moving computer vision to the edge. The integration of AWS Panorama into TensorIoT’s SafetyVisor product, our general solutions, and our retail intelligence gives TensorIoT the flexibility to customize solutions even further for clients looking to harness the power of edge computer vision.
For a practical example of this tool in action, today we will walk you through the steps for setting up AWS Panorama for detecting and counting people.
Step 1: Sign-In into the AWS Console and select AWS Panorama from the services menu.
Step 2: In the Getting Started Page, Select Step 1: Set up the appliance.
Step 3: Begin Setup by naming your appliance (Edge Hardware Device).
Step 4: Configure networking for your appliance
There are two options:
- WIFI
- Ethernet
For our example, we select WIFI and enter the Wifi information in the required fields.
Step 5: SSH Step
You can Activate or Deactivate SSH for your Appliance.
Step 6: Download the configuration file in the USB stick.
Make sure the USB stick is formatted with FAT32 before putting the downloaded zip file in the USB stick.
Step 7: Connect the USB drive, and power cable to start your appliance
Once the process is complete, the device will show online in the console.
In a follow-up blog, we’ll explore connecting an iPhone or any other camera using RTSP stream and creating an application to detect and count the people. Stay tuned!
About us —
As a launch partner for AWS Panorama, TensorIoT has invested in training, updating models, and integrating Panorama into our Computer Vision Solutions. AWS Panorama brings Cloud-to-Edge to a managed service for creating, deploying, and managing applications at the Edge.
