Video Frame Analysis via Captured Image Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video search methods rely on text-based meta information, failing to recognize popular scenes or discussions about videos on the internet, especially when specific video scenes are shared instead of the entire video, and lack connection of rich meta information with captured images.
Innovation Solution
An apparatus and method that extracts frames from videos, digitizes their features, compares them with captured images from external sources, and maps related information to corresponding frames, providing meta information and analysis on video popularity through social rankings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If text-based meta information is used for video search, then search simplicity is maintained, but the ability to recognize popular scenes and discussions about videos is lost
Solution Approach 1:
The patent segments video content into individual frames and compares them with captured images from external sources. By dividing the video into discrete frame units that can be independently matched with screenshots, the system enables scene-specific recognition while maintaining overall search functionality. This segmentation allows popular scenes to be identified without requiring complex full-video analysis.
Solution Approach 2:
The patent introduces captured images (screenshots) as an intermediary element between video content and user search queries. These captured images serve as mediators that bridge the gap between text-based search and visual content recognition. By comparing extracted video frames with these intermediary image elements, the system can identify popular scenes and discussions without abandoning simple text-based search interfaces.
2Loss of information
If captured images from external sources are collected and compared with video frames, then rich meta information and scene analysis are provided, but system complexity increases
Solution Approach 1:
The patent performs preliminary actions by pre-collecting and storing captured images from external sources before the actual video analysis process. These pre-collected images are stored in a database, ready for comparison. This preliminary preparation reduces the complexity of the main analysis process, as the system only needs to perform frame extraction and comparison operations rather than gathering all data during the analysis phase.
Solution Approach 2:
The patent uses captured images (copies of video screenshots) as simplified representations of the actual video content. Instead of analyzing entire videos or complex metadata structures, the system works with these image copies that capture essential visual information. This copying approach simplifies the data structure while preserving the ability to perform meaningful comparisons and extract rich meta information.
3Ease of operation
If specific video scenes are shared instead of entire videos, then ease of discussion is improved, but connection of rich meta information with captured images is lost
Solution Approach 1:
The patent implements feedback mechanisms by comparing extracted video frames with captured images and using the match results to provide enhanced meta information. When a captured image matches a video frame, the system provides feedback that includes not only the match confirmation but also rich contextual information about the video, such as metadata, timestamps, and related content. This feedback loop maintains discussion ease while restoring the connection to rich meta information.
Data Source
AI summary
An apparatus for analyzing a video capture screen includes: a video frame extracting unit extracting at least one frame from a video having a plurality of frames; an extracted frame digitizing unit digitizing features of each of the at least one frame extracted by the video frame extracting unit; an image digitizing unit digitizing features of at least one collected search target image; an image comparing and searching unit comparing the search target image with the at least one frame extracted from the plurality of frames by digitized values of the collected search target image and the at least one frame; and a search result processing unit mapping related information of the collected search target image to a frame coinciding with the search target image and storing the related information in a database, when the extracted at least one frame coinciding with the search target image is present in a comparison result.