Video Content Template Generation for Piracy Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting pirated video content are inefficient in identifying fuzzy copies, which are modified versions of original content, as they rely on metadata analysis that can be easily manipulated, leading to high computational loads and reduced accuracy.
Innovation Solution
A method that creates a template of original video content by extracting metadata and frame sequences, generating a vector of scenes, and comparing it with potential copies to identify fuzzy duplicates, reducing computational burden and improving detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If metadata analysis is used to detect pirated video content, then the detection process can be automated, but the accuracy deteriorates because metadata can be easily manipulated and changed by pirates
Solution Approach 1:
The patent extracts only the most reliable and difficult-to-manipulate metadata fields (such as frame dimensions, color depth, and encoding parameters) while discarding or weighing less reliable fields (such as title, description, and uploader information). This selective extraction maintains automation while improving accuracy by focusing on metadata that is harder for pirates to alter.
Solution Approach 2:
The patent changes the approach from using traditional metadata fields to analyzing modified metadata fields that are harder to manipulate. By transforming the detection focus to parameters like frame dimensions, color depth, and encoding characteristics, the system achieves both automation and improved accuracy against metadata manipulation.
2Measurement precision
If the entire video file is downloaded and analyzed to identify fuzzy copies, then the detection accuracy improves, but the computational load and data transmission requirements increase significantly
Solution Approach 1:
The patent extracts only the essential metadata fields needed for detection (frame dimensions, color depth, encoding parameters, and temporal information) rather than downloading and analyzing the entire video file. This extraction approach maintains sufficient detection accuracy while dramatically reducing computational load and data transmission requirements.
Solution Approach 2:
The patent segments the video analysis task by focusing only on specific metadata parameters rather than processing the complete video content. By dividing the detection task into analyzing individual metadata fields separately, the system achieves accurate fuzzy copy detection without requiring full file download and processing.
3Device complexity
If traditional algorithms are used to detect pirated content, then the initial implementation is simpler, but the reliability deteriorates as pirates use tools to bypass detection by changing video content
Solution Approach 1:
The patent changes the detection parameters from simple file-based metadata to more robust visual and technical parameters (frame dimensions, color depth, encoding characteristics). This parameter transformation increases reliability by detecting changes that are harder for pirates to manipulate, while the systematic approach keeps the implementation manageable.
Solution Approach 2:
The patent performs preliminary analysis by pre-defining the set of metadata fields to extract and the comparison criteria before actual detection runs. This preliminary preparation includes identifying which metadata fields are most reliable and establishing the comparison methodology in advance, which improves reliability while keeping the actual detection process systematic and manageable.
Data Source
AI summary
There is disclosed a method of creating a template of original video content, which is performed on a computer device that has access to a previously generated database of original video content. The method comprises receiving identifiers for at least a portion of an original video content; extracting at least a portion of metadata of the original video content; extracting at least a portion of frames from a sequence of frames of the original video content; identifying a sequence of scenes; creating a vector of the sequence of scenes; generating a template of the original video content that includes at least the portion of the metadata, and a vector of the sequence of scenes of the original video content; and storing the template in a database.