Video Sequence Comparison via Fingerprint Temporal Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current monitoring solutions for video workflows cannot accurately compare video sequences with different image rates across multiple test points without human intervention, as existing systems are invasive or only suitable for offline use.
Innovation Solution
A method is developed to compare video fingerprints by generating a converted sequence of fingerprints at a different image repetition rate, using spatial or temporal fingerprints, and performing correlation between sequences, which allows for dynamic and non-invasive monitoring across various video standards.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If existing fingerprinting systems are used to compare video sequences, then video content can be characterized with low-bandwidth representation, but the systems can only handle video standards with the same image rate or require invasive test stimuli insertion
Solution Approach 1:
The patent transforms temporal fingerprints from one sampling rate to another by changing the temporal parameter (image repetition rate) while preserving the essential content characteristics. This allows fingerprints generated at different image rates to be compared directly without requiring the video sequences to match in their original temporal characteristics
Solution Approach 2:
The patent introduces an intermediary transformation process that converts fingerprints from different temporal domains into a common reference frame. This intermediary step enables comparison between fingerprints from video sequences with different image rates without direct interaction between the original sequences
2Reliability
If invasive test stimuli insertion is used to enable comparison between different video standards, then monitoring can be performed, but the process becomes invasive and suitable only for offline set-up scenarios
Solution Approach 1:
The patent extracts the essential content characteristics into temporal fingerprints that can be independently analyzed and transformed. By separating the content representation from the original video stream, the system can perform comparisons without inserting test stimuli into the actual video content
Solution Approach 2:
The system uses the existing video content itself to generate fingerprints for comparison, rather than requiring external test stimuli. The video sequences self-provide the necessary information for monitoring through their own temporal and spatial characteristics
3Productivity
If existing systems are used for dynamic monitoring, then real-time comparison is needed, but no industry monitoring solution is available for comparing different video image rates across multiple test points
Solution Approach 1:
The patent creates a universal comparison mechanism that can handle multiple video standards with different image rates through a single temporal transformation approach. The system can process and compare fingerprints from various video sources (different standards, different rates) using the same reference frame methodology
Solution Approach 2:
The patent performs preliminary transformation of fingerprints into a common reference frame before comparison. By pre-processing the temporal characteristics of fingerprints from different video sequences, the system prepares them for direct comparison without requiring complex real-time adjustment during the monitoring process
Data Source
AI summary
A first sequence of video fingerprints at a first image repetition rate is compared with a second sequence of video fingerprints at a second, different image repetition rate. A converted sequence of fingerprints is generated from the first or second sequence of fingerprints by forming a combination of the values of a number of neighbouring fingerprints. The combination is a weighted average based on the temporal location of the respective neighbouring fingerprints in the first sequence of fingerprints with respect to the fingerprint in the converted sequence. A correlation is then performed between the converted sequence of fingerprints and the other sequence of fingerprints.