Video Scene Verification Using Typical Scene Characteristic Amounts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video data verification technologies require referencing all thumbnail images, limiting the reduction of verification time and resource efficiency.
Innovation Solution
A video processing system that stores and associates scenes and their characteristic amounts, defining one scene as typical and others as derived, allowing for rapid verification of new scenes by comparing their characteristic amounts with the typical scene.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all thumbnail images are referenced for video data verification, then verification accuracy is maintained, but verification time increases and processing efficiency decreases
Solution Approach 1:
The patent segments the verification process into two stages: first verifying with a small number of key frames (segmented from the video), then selectively verifying additional frames only when needed. This segmentation allows the system to achieve high verification accuracy while significantly reducing the time required compared to verifying all frames.
Solution Approach 2:
The patent performs preliminary verification using a small set of key frames before determining whether full verification is necessary. This preliminary action allows the system to quickly filter out cases that don't require extensive verification, reducing overall verification time while maintaining accuracy for cases that do require thorough checking.
2Reliability
If all thumbnail images are referenced for video data verification, then comprehensive verification is achieved, but resource consumption increases
Solution Approach 1:
The patent applies partial action by verifying only the necessary portion of video frames rather than all frames. The system determines the appropriate verification depth based on initial assessment, consuming only the resources needed to achieve reliable verification results without the excessive resource consumption of full verification.
Solution Approach 2:
The patent changes the verification parameter from a fixed approach (verifying all frames) to a dynamic approach where the verification depth is adjusted based on the specific video content and initial assessment results. This parameter change allows the system to maintain high reliability while optimizing resource consumption for each verification task.
3Measurement precision
If traditional verification methods are used, then all video frames are checked, but processing speed is limited
Solution Approach 1:
The patent segments video verification into hierarchical levels: key frame verification, scene verification, and full frame verification. This segmentation enables the system to process multiple videos in parallel at different levels, significantly increasing verification throughput while maintaining thoroughness for videos that require it.
Solution Approach 2:
The patent implements periodic verification where videos are first checked at a fast pace using key frames, then additional verification is periodically applied only to videos that fail the initial check. This periodic action pattern increases overall verification throughput by processing most videos quickly while still maintaining thorough verification standards.
Data Source
AI summary
A video processing apparatus includes a storage unit, an input unit, and a determining unit. The storage unit stores a plurality of scenes included in a scene group and characteristic amounts which are extracted from a series of plural frames included in the respective scenes in the scene group, in association with each other, defining one scene out of the plurality of scenes included in the scene group as a typical scene and the other scenes as a derived scene, the scene group being a group of a plurality of scenes derived from a common scene. The input unit inputs a characteristic amount which is extracted from each of the frames of a new scene. The determining unit determines whether the new scene represented by the inputted characteristic amount is the derived scene of the scene group or not by verifying the inputted characteristic amount with the characteristic amount of the typical scene of the scene group stored in the storage unit.