Computer Vision News - April 2018

This month we shall check out openCV Computer Vision library on Amazon Web Services (AWS)’s Lambda cloud service. AWS Lambda allows you to remotely execute and run your dedicated code on Amazon’s cloud servers. All this without the need for buying or maintaining servers. One of the nice things about this AWS Lambda service is that you only pay for the time the code actually runs. The service allows you to run code for almost any application conceivable. It is very simple to use, all you need to do is upload the code to AWS-Lambda (will show you in a second how) and the servers will take care of all maintenance needed. Additionally, you have the capability to define automatic execution of the code from other AWS services, or running it directly from any Internet application or mobile app. The service has a number of advantages as far as computer vision applications are concerned, including: 1) when local computational power is low (for instance, a mobile device with weak processor); 2) when you want to keep your algorithm confidential; or 3) when it’s not possible to install needed computer vision software packages on the dedicated computer. We will demonstrate and explain how to create Lambda functions that support OpenCV on AWS and call them remotely. The process consists of the following steps: 1. Opening an AWS.Amazon account. 2. Setting up a Python work environment that includes OpenCV. 3. Writing your code. 4. Uploading your code to the Lambda service. 5. Ongoing remote code development using the Cloud9 work environment. 6. Linking the code to an application or its automatic execution. 12 We Tried for You Tool by Assaf Spanier “The service allows you to run code for almost any application conceivable… …you only pay for the time the code actually runs.” Computer Vision News Computer Vision on Amazon Cloud