Video Graph Sequence Extraction for Qualitative Feature Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional video understanding technologies are limited in identifying and quantitatively extracting qualitative factors such as social, cultural, and artistic features that significantly impact viewer satisfaction and immersion, failing to effectively analyze and utilize these elements in video metadata tagging.
Innovation Solution
A method involving graph embedding technology and deep neural networks to store video data in a graph structure, segmenting videos into segments, generating video graph sequences based on object and edge feature vectors, and classifying these sequences to extract and classify qualitative characteristics, allowing for the identification of video styles and storage in a database.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional video metadata tagging technology is used, then basic video information can be recognized, but qualitative characteristics such as social, cultural, and artistic features cannot be effectively extracted
Solution Approach 1:
The video is segmented into multiple video segments, and each segment is processed independently to extract qualitative characteristics. This segmentation allows the system to handle complex video content in manageable units, improving extraction precision without overwhelming the system
Solution Approach 2:
A graph structure is introduced as an intermediary representation between raw video data and qualitative characteristic extraction. The graph embedding technology creates intermediate representations that capture semantic relationships, enabling precise extraction of social, cultural, and artistic features while managing system complexity through structured intermediate processing
2Measurement precision
If graph embedding technology and deep neural networks are used to extract qualitative characteristics, then extraction precision improves, but processing time and computational resources increase
Solution Approach 1:
By segmenting the video into smaller units and processing them independently through graph embedding and deep neural networks, the system achieves high extraction precision while enabling parallel processing that reduces overall processing time compared to analyzing the entire video sequentially
Solution Approach 2:
The system extracts only the necessary qualitative characteristics (social, cultural, artistic features) rather than processing all possible video attributes. This selective extraction approach maintains high precision for target features while reducing unnecessary computational overhead and processing time
Data Source
AI summary
A method, a device and a recording medium for extracting a qualitative characteristic of a video may include segmenting a video into at least one video segment, and based on an edge representing a relationship between objects of the video segment and the objects, generating video graph sequences in a graph form of the video segment, and generation of the video graph sequences may be performed based on an object feature vector expressing a qualitative characteristic of the objects as a vector and an edge feature vector expressing a qualitative characteristic of the edge as a vector.