What does this Amazon Web Services Solution do?
The process of transforming, sharpening, upscaling low resolution video into high resolution video is called Super Resolution, for instance, upscale 480p video to 4K (Ultra-HD) video. Amazon Web Services offers a deep learning algorithm for video super resolution task. The algorithm helps you improve the quality of your videos with low cost. The solution allows you to quickly deploy the algorithm and corresponding architecture into your own account.
Amazon Web Services Solution overview
This solution provides a CloudFormation template allowing you to deploy all the necessary cloud resources into your account with just one click. The architecture is shown below.
Amazon API Gateway automatically triggers a Amazon Lambda function to read the source video from S3, cuts the video into several segments, and stores them in Amazon Elastic File System (Amazon EFS). The Lambda function then submits two batch processing jobs to Amazon Batch: one is "video super-resolution job" and the other is "video merging job". The Amazon Batch uses the Amazon EC2 Inf1 instances on Amazon Elastic Container Service (Amazon ECS) to run the jobs.
- The "video super-resolution job" firstly pulls the video super-resolution ECR image from Amazon Elastic Container Registry (Amazon ECR), and retreives all the video segments from Amazon EFS, then applies the "Super Resolution" algorithm on each of the segments. At the end, the job saves the processed video segments back into Amazon EFS.
- The "video merging job" retrieves all the processed video segments from Amazon EFS, merges them back together, and puts the final video into Amazon S3.
AI Video Super Resolution
Last updated: 09/2022
Author: Amazon Web Services
Estimated deployment time: 5 min
Browse our portfolio of Amazon Web Services-built solutions to common architectural problems.
Find Amazon Web Services certified consulting and technology partners to help you get started.
Sign-up and start exploring our services.