Frame-based Analysis for Your Videos
Frame-based Analysis for Your Videos


Amazon Web Services (AWS) offers powerful and cost-effective services to help customers process, analyze, and extract meaningful data from their video files. Customers who want to obtain a broader understanding of their video libraries can use AWS services to develop solutions that quickly and seamlessly analyze frames in their video files. However, developing these solutions can sometimes require extensive knowledge of deep-learning algorithms, depending on the level of analysis required. 

To help customers more easily search, query, and analyze their video libraries, AWS offers the Frame-based Analysis for Your Videos solution. This solution combines Amazon Rekognition with the open source software FFmpeg, a fast video converter, to identify and extract frame-level metadata from video files. This AWS solution automatically detects and extracts video frames using machine learning algorithms to identify facial, object, and scene-level metadata in those frames, and stores the resulting metadata for future search and analysis. It also includes a facial search feature that uses customer-provided image files to search for matches in the collected video metadata.


You are responsible for the cost of the AWS services used while running this reference deployment. As of the date of publication, the estimated base Amazon Elastic Compute Cloud (Amazon EC2) and Amazon DynamoDB capacity costs for running this solution with default settings in the US East (N. Virginia) Region is $0.22 per hour. This cost estimate can scale up to $1.12 per hour based on the number of simultaneous videos being processed.

Video frame processing costs will vary based on the number of frames analyzed. This solution processes a frame every second, which will result in 3,600 frame images per hour of video. The solution uses Amazon Rekognition to process the subset of images that contain faces. By default, this solution processes frames every second, which will result in 3,600 frame images per hour of video. Each frame is processed by the Amazon Rekognition IndexFaces and DetectLabels API calls, resulting in two Amazon Rekognition API calls per frame image.

See the  Amazon Rekognition pricing page for more information, including details about free tier and volume discount pricing. Note that this estimate does not include charges incurred from Amazon Simple Storage Service (Amazon S3) and Amazon DynamoDB storage, AWS IoT, Amazon Simple Queue Service (Amazon SQS), or data transfer costs. Prices are subject to change. For full details, see the pricing webpage for each AWS service you will be using in this solution.

On this page: