Machine Learning

Building Video Content Moderation with Amazon Rekognition - Ensuring Safety on UGC Platforms

Automate video moderation on UGC platforms, and streamline media workflows with face search and segment detection. Learn how to build event-driven pipelines with S3 and Lambda.

約 3 分で読めます

Rekognition Video API Overview

Rekognition's Video API asynchronously analyzes video files stored in S3. You initiate analysis with Start APIs such as StartContentModeration, StartLabelDetection, and StartFaceSearch, and a completion notification is sent to an SNS topic. After completion, you retrieve results using the corresponding Get API. Results include per-frame timestamps, allowing you to pinpoint exactly when something was detected in the video. Supported video formats are H.264-encoded MOV and MP4, with a maximum file size of 10 GB and a maximum video length of 6 hours.

Implementing Content Moderation

UGC (user-generated content) platforms need to automatically check whether user-uploaded videos contain inappropriate content. You can build a pipeline that triggers a Lambda function on S3 video upload to call StartContentModeration, detects completion via SNS notification, and evaluates the results. Detection results include inappropriate content categories (Explicit Nudity, Violence, Visually Disturbing, etc.) and confidence scores. Set a confidence threshold and design a workflow that automatically unpublishes videos exceeding the threshold or routes them to a human review queue.

Face Search and Segment Detection

By pre-registering face images in a face collection, StartFaceSearch can detect scenes in a video where registered faces appear. This is useful for media companies searching for scenes featuring specific performers, or for security applications tracking specific individuals. Segment detection (StartSegmentDetection) analyzes video structure, automatically identifying shot boundaries (camera cuts), black frames, color bars, slates, studio logos, and end credits. It's effective for automating video editing at broadcast stations and streaming services, and for detecting ad insertion points. To understand content moderation model design, related books (Amazon) are a helpful reference.

Rekognition Video Pricing

Rekognition Video API pricing is based on the minutes of video processed. Content moderation costs approximately $0.12 per minute, label detection approximately $0.10 per minute, and face search approximately $0.10 per minute. A 10-minute video costs about $1.20 for content moderation. For UGC platforms processing large volumes of video, optimizations such as setting appropriate confidence thresholds to reduce unnecessary analysis and substituting the Image API for thumbnail analysis on short videos (under 30 seconds) are effective.

Summary

Rekognition Video API enables serverless automated video content analysis. Content moderation ensures safety on UGC platforms, while face search and segment detection automate media workflows. By combining S3 and Lambda in an event-driven architecture, you can fully automate the pipeline from video upload to analysis completion.

Related Services

Amazon RekognitionAn ML-based image recognition service that makes image and video analysis easy via APIAmazon S3A scalable object storage serviceAmazon KinesisA service for collecting, processing, and analyzing real-time streaming data

Related Articles

Implementing Image and Video Analysis with Amazon Rekognition - From Label Detection to Custom ModelsLearn how to implement label detection, facial analysis, and text detection using pre-trained APIs, and build domain-specific image recognition models with Custom Labels.Building a Real-Time Data Pipeline with Amazon Kinesis - Choosing Between Data Streams and Data FirehoseIngest real-time data with Data Streams and automatically deliver it to S3, Redshift, and OpenSearch with Data Firehose. This article explains how to build a streaming pipeline with shard design and on-demand mode selection.Building an Apache Kafka Streaming Platform with Amazon MSK - Cluster Design and OperationsDesign a managed Apache Kafka cluster, learn when to choose MSK Serverless, and explore data integration patterns with MSK Connect.

More on This Topic

GPU-Based Machine Learning Training with AWS Batch - Cost-Efficient Large-Scale TrainingRun GPU training with your existing Docker containers, and cut costs by up to 90% using Spot Instances and checkpointing. Includes guidance on when to choose Batch over SageMaker.Using Claude on Amazon Bedrock - Model Selection, Prompt Design, and Cost OptimizationCompares the Anthropic Claude models available on Amazon Bedrock, provides model selection guidelines by use case, and covers prompt design best practices and cost optimization.Building RAG Applications with Amazon Bedrock Knowledge Bases - Implementing Retrieval-Augmented GenerationAutomatically index documents on S3 and unify search and generation with the RetrieveAndGenerate API. Covers chunking strategy selection and safety enforcement with Guardrails.Getting Started with Quantum Computing on Amazon Braket - Designing and Simulating Quantum CircuitsPrototype for free with local simulators, then run quantum circuits on IonQ and Rigetti hardware. Covers implementing VQE and QAOA with hybrid jobs.Privacy-Preserving ML with AWS Clean Rooms ML - Build Models Without Sharing DataLearn how to build lookalike models with Clean Rooms ML, apply differential privacy, and leverage the results for ad targeting.Implementing Natural Language Processing with Amazon Comprehend - Sentiment Analysis and Entity ExtractionLearn about sentiment analysis, entity extraction, and building custom classification models with Comprehend.Building Conversational Bots - Natural Conversation Interfaces with Amazon Lex and PollyLearn how to build conversational bots using Amazon Lex and Amazon Polly.Demand Forecasting - Predicting the Future from Time Series Data with Amazon ForecastInput historical time series data and related variables to automatically build ML-based demand forecasting models. This guide covers forecast accuracy evaluation metrics and patterns for leveraging forecast results through S3 and QuickSight integration.

Similar Articles and Services

Amazon RekognitionA machine learning-based visual analysis service that performs object detection, facial analysis, text detection, content moderation, and custom label detection on images and videosImplementing Image and Video Analysis with Amazon Rekognition - From Label Detection to Custom ModelsLearn how to implement label detection, facial analysis, and text detection using pre-trained APIs, and build domain-specific image recognition models with Custom Labels.Video Transcoding - Building a Scalable Media Conversion Pipeline with AWS Elemental MediaConvertLearn how to build a video transcoding pipeline using AWS Elemental MediaConvert and S3. This article covers multi-format output, HDR support, and cost-efficient serverless media processing techniques.Serverless Video Transcoding with AWS Elemental MediaConvert - HLS Delivery and Thumbnail GenerationTranscode file-based video at broadcast quality and generate multiple formats including HLS, DASH, and MP4 in a single job. Learn how to build large-scale video processing pipelines with job templates and queue priority control.