Video Signature Generation Using Compressed Intra-Coded Frames
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video signature generation systems for identifying digital video information are inefficient due to reliance on uncompressed data, leading to high costs and slow processing speeds, and are often aspect ratio and frame rate dependent, resulting in increased false match rates and failure to identify programs.
Innovation Solution
The system generates video signatures using compressed digital video information, specifically selecting intra-coded frames and extracting DC coefficient information to calculate center of mass or centroid values, which are then used to create unique signatures for video programs, independent of image format and frame rate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If uncompressed video data is used to generate signatures, then signature accuracy is improved, but processing cost and hardware requirements increase significantly
Solution Approach 1:
The patent extracts only the essential signature information from video data by using compressed video frames and selecting specific intra-coded blocks, rather than processing all uncompressed video data. This extraction approach maintains sufficient signature accuracy while dramatically reducing processing complexity and hardware requirements.
Solution Approach 2:
The patent changes the processing parameters from uncompressed video data to compressed video data, specifically utilizing intra-coded frames and blocks from MPEG-compressed video streams. This parameter change allows signature generation to work effectively with compressed data, reducing the resource requirements while maintaining identification accuracy.
2Device complexity
If software-based signature generation is used, then device complexity is reduced, but processing speed becomes too slow to capture important information
Solution Approach 1:
The patent extracts only the necessary signature data from selectively processed compressed video frames, focusing computation on essential intra-coded blocks rather than entire frames. This selective extraction enables software-based systems to achieve adequate processing speed without requiring expensive hardware acceleration.
Solution Approach 2:
The patent applies partial processing by selecting only certain intra-coded frames and blocks within frames for signature generation, rather than processing all video data. This partial action approach provides sufficient signature information at a manageable software processing speed.
3Productivity
If less video information is used to generate signatures, then processing speed is improved, but false match rate increases and reliability decreases
Solution Approach 1:
The patent applies local quality by selecting specific intra-coded blocks within video frames that contain the most distinctive visual information for signature generation. This localized approach ensures that the reduced data set still captures sufficient detail for reliable identification, maintaining accuracy while improving processing speed.
Solution Approach 2:
The patent changes from using all video information to using selectively processed compressed video data, specifically leveraging intra-coded frames and blocks. This parameter change optimizes the balance between processing speed and reliability by using the minimum necessary information from compressed sources.
4Ease of manufacture
If traditional signature systems are used, then they are simple to implement, but they are dependent on specific image formats and frame rates, reducing versatility
Solution Approach 1:
The patent achieves universality by designing a signature generation system that works with compressed video data from multiple sources and formats. By processing intra-coded frames and blocks from MPEG-compressed video streams, the system becomes adaptable to different aspect ratios, frame rates, and video sources while maintaining consistent identification performance.
Solution Approach 2:
The patent changes the input parameters from fixed-format uncompressed video to flexible compressed video data structures. This allows the system to accept various image formats and frame rates while maintaining signature generation capability, significantly improving versatility without complicating the core processing architecture.
Data Source
AI summary
Digital video signature apparatus and methods for use with video program identification systems are disclosed. The disclosed apparatus and methods identify a video program using a sequence of signatures. Each of the signatures includes a set of centroids corresponding to one of a plurality of frames of the video program. The apparatus and methods compare the sequence of signatures to a set of reference sequences of signatures and identify the video program based on the comparison of the sequence of signatures to the set of reference sequences of signatures.