Video Deduplication
Media & EntertainmentBusiness Impacts
Audit Report
Time and Effort Minimization
Cost Reduction
Customer Key Facts
- Location : North America
- Industry : Media and Entertainment
Problem Context
The client is one of the world’s leading media and entertainment companies in the development, production, and marketing of entertainment, news, and information to a global audience. When a video file moves from production to broadcast or publishing department, it undergoes commercial and/or compliance changes to adhere to a different market. At each stage, multiple duplicates are created which results in huge archival size and retrieval cost.
Challenges
- Huge archived data to process
- Multiple data sources in different formats
- Restore selected assets by the user in their native specifications
Technologies Used
Amazon EKS
Amazon CloudWatch
AWS SageMaker
Amazon S3
Amazon IAM
Amazon EC2
AWS CloudFormation
An automated solution to identify and delete duplicate audio and video content form archive storage
Solution
Quantiphi built an automated solution to identify and delete duplicate audio and video content stored in their archive. This AI solution generates high-quality embeddings using audio and computer vision-based feature extractors to identify unique logical segments in the content’s video, audio, and subtitle tracks. The solution showcases the ability to restore selected assets by the user in its native specifications.
Result
- The AI-based solution saved a significant cost & time incurred by manually identifying and deleting duplicate audio-visual content
- Delete & Restore ~5% deleted assets of the customer’s choice
- Benchmark the duplicate segment identification process on 5% of the sample set
- Deliver an audit report with a comprehensive summary of the deduplication exercise
