:Media fingerprinting

Scenarios

• Original content identification: Accurately identifies whether a video is original, edited, or recreated by we media. This provides technical support for copyright protection and helps build a healthy video ecosystem where the rights of video producers and short video platforms are protected.

• Duplicate video identification: Creates a media fingerprint library to compare video fingerprints with those in the library. This helps identify duplicate videos and duplicate clips. This prevents excessive identical or similar videos from being pushed to users and affecting user experience during content delivery and personalized recommendations.

• Rapid content review: Maintains a library of prohibited videos or a custom blacklist to quickly compare DNA fingerprints of new videos and determine whether they contain prohibited content. Compared with traditional review methods, this improves the efficiency of reviewing large volumes of videos, ensures accuracy, and reduces review costs.

• Ad revenue sharing: Uses video DNA technology to retrieve ads from the revenue sharing ad library and monitor and identify specific ads. This facilitates ad revenue sharing in real time and provides an easier method to specify the time and frequency to deliver ads. This creates more value for ad producers and advertising platforms.

  Note

  Media DNA currently only supports detecting known advertisement content in the DNA library. It cannot detect advertisement content that has not been added to the library. To detect advertisements, you must first upload the advertisement materials to the DNA library.

Preparations

Perform the following operations before you create a media fingerprinting job:

• Call the CreateDNADB operation as needed to create a DNA library.

• Upload resources: Upload the media resources to be processed to OSS or the media library.

• Configure callbacks (optional): Receive callbacks when jobs are completed to query job details. For more information, see Configure callbacks.

Submit media DNA jobs through OpenAPI

1. Create a media DNA library.

   Submit a task to create a media DNA library. The system returns information about the new DNA library. After the DNA library is created, its status is updated to active. When you submit the job, provide the UID of your Alibaba Cloud account. After the media fingerprint library is created, bind the UID of your Alibaba Cloud account to the media fingerprint library. For more information, see CreateDNADB.

   Note

   A media DNA library is used to record and store media DNA. The media DNA service supports the following features:

   • Creating media DNA libraries by user (UID).

   • Creating multiple media DNA libraries for different business scenarios for the same user (UID).

2. Submit a media DNA job.

   • The SubmitDNAJob operation supports processing real-time, incremental, and existing videos. You can use this operation to submit DNA processing tasks for various types of media.

   • For video DNA jobs, if you have fewer than 2,000 hours of videos, you can directly submit tasks through the SubmitDNAJob operation. If you have many base videos, submit a ticket to request the corresponding resources.

3. Query processing results.

   • Query job results through the API: You can use QueryDNAJobList.

   • Query job results through callbacks: See Set callback events.

   • The processing results contain the following main information.

     • Whether the job has found duplicate media.

     • If duplicate media is found, the system outputs the duplicate media information.

     • The degree of duplication of the media (the duplication degree is a value between 0 and 1).