Amazon Rekognition Video review
A usage-priced AWS video analysis API with broad detection features and deep integrations.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
Amazon Rekognition Video is a managed AWS machine-learning service for analyzing video stored in Amazon S3 and, for eligible existing customers, live streams through Amazon Kinesis Video Streams. It is aimed at teams building media indexing, connected-home monitoring, retail analytics, public-safety, and brand-safety workflows. The service detects objects, scenes, activities, landmarks, people, faces, celebrities, text, and inappropriate content, with timestamps and selected bounding-box data. It supports both asynchronous stored-video analysis and real-time streaming events through cloud APIs.
Pricing follows a usage-based model rather than fixed subscription tiers. The published Usage-based plan starts at $0.00817 per minute and charges only for processed video. Pricing varies by feature and region, so the total cost depends on the analysis capabilities and video volume involved. There is no free plan. This structure fits projects that prefer variable costs tied to processing, while teams seeking a complete video-operations application may need additional AWS services or another product. The service is available through web and API access, with documentation listed as the support channel.
Its strongest fit is the AWS ecosystem: integrations include Amazon S3, Amazon Kinesis Video Streams, Amazon SNS, Amazon SQS, AWS Lambda, Amazon Transcribe, Media2Cloud, and Media Insights Engine. Features include person tracking and pathing, text detection with timestamps and bounding boxes, face search, content moderation, and real-time alerts. However, AWS states that Streaming Video Analysis is unavailable to new customers effective April 30, 2026; eligible existing customers retain access, while other Rekognition features remain available. Choose Amazon Rekognition Video for cloud-based, usage-priced analysis within AWS. Choose an alternative if you need a packaged operations platform or newly provisioned streaming analysis.
Amazon Rekognition Video pros and cons
- Where it wins
- Pay only for processed video at usage-based rates
- Analyzes objects, people, faces, text, activities, and content
- Integrates with S3, Kinesis Video Streams, Lambda, and AWS messaging
- Where it doesn't
- No free plan is available
- It is primarily an API, not a complete video-operations platform
- Streaming Video Analysis is unavailable to new customers after April 30, 2026
Amazon Rekognition Video fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update