Our technologies
Rapid and precise search and identification
WebKontrol provides video recognition and fingerprint matching software that automatically prevents the upload of unauthorized videos to platforms, builds a transparent monetization system for creators, and improves video ads’ performance. WebKontrol helps platforms grow user engagement and drive down the cost of content moderation. The product has been designed to enable platforms with user-generated content to comply with Article 17 of the EU Copyright Directive.
The principle of automatic video identification is similar to that of YouTube Content ID. Every piece of content uploaded by platform users is automatically checked against rightsholders' fingerprint database.
WebKontrol’s video fingerprinting and matching technology enables:
- automated identification of infringing content (i.e. copyrighted, and other), so platforms can avoid copyright claims and legal proceedings;
- enhanced monetization and statistics tracking for content owners, which increases their loyalty to a platform;
- automated pre-filtering of illegal videos prior to their public access to protect platform brand and viewers;
- detection of duplicate videos inside a selected video database.
The rapid detection of all copies of a video file is achieved regardless of the video’s quality, audio, and different distortions. The entire technology behind WebKontrol’s video search engine has been developed by WebKontrol engineers and has been patented.
What makes our search truly efficient?
Cost-effectiveness. Automation of video pre-filtering eliminates the need for extensive teams of content moderators
Our reference database. It is steadily growing; the index of major hosting, streaming, and file-sharing sites are constantly updated, and content identification takes place 24/7. This ensures rapid response rates once matches are detected.
Scale. Matching can be performed simultaneously for hundreds of thousands of videos
Speed. Matching results are returned in a matter of seconds
High accuracy. Nearly-zero probability of false positive match results
In-depth search. All copies are detected based on video content, regardless of audio and file name used on a platform. This enables to perform an in-depth search and identify purposefully misnamed files, such as “123” etc
