Dynamic Relationship Graph for Video Storyline Visualization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users of long-term video series face difficulties in recalling the current relationships between characters due to the lack of comprehensive summaries, especially during off-seasons when months pass without new episodes, and existing summaries often fail to cover all relationships.
Innovation Solution
A dynamic relationship visualization system that displays relationship diagrams associated with each portion of the video, using nodes and edges to represent characters and their relationships, which change over time, allowing users to explore and customize the presentation of relationships through various diagram types like relationship graphs, partial graphs, and emotion curves.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If users rely on existing summaries for long-term video series, then some relationship information is provided, but the summaries fail to cover all relationships and are unavailable during off-seasons
Solution Approach 1:
The system pre-processes the video content to extract and store character relationship information in advance, creating a comprehensive relationship graph that can be queried at any time. This preliminary action ensures that relationship information is always available, eliminating the dependency on periodic summaries and covering all relationships throughout the series.
Solution Approach 2:
The patent introduces an intermediary relationship graph structure that mediates between the video content and the user. This graph serves as a persistent knowledge base that translates complex video narrative into structured relationship data, making comprehensive relationship information continuously accessible regardless of whether new episodes are released.
2Loss of information
If comprehensive summaries are provided for all relationships, then complete information is available, but the complexity of tracking and presenting all relationships increases significantly
Solution Approach 1:
The system segments the comprehensive relationship information into discrete, manageable units represented as nodes and edges in a relationship graph. Each character becomes a node, and each relationship becomes an edge with specific attributes. This segmentation allows the system to handle complex relationships in a structured way while providing users with flexible querying capabilities to focus on specific relationships of interest.
Solution Approach 2:
The relationship graph is designed to be dynamic, allowing relationships to be added, removed, or modified as the video series progresses. The system can adaptively update the graph structure and provide different visualization levels, from individual relationship details to comprehensive relationship networks, enabling users to manage information complexity according to their needs.
3Loss of information
If detailed relationship information is provided continuously, then users have complete understanding, but it may overwhelm users during video playback
Solution Approach 1:
The system implements partial action by providing relationship information on-demand rather than continuously. Users can query the relationship graph at specific moments during playback to obtain relevant relationship context, receiving only the information they need at that moment. This approach maintains complete underlying information while presenting it in a controlled, non-overwhelming manner during video playback.
Data Source
AI summary
A computer-implemented method includes determining that a first time point in a video has been reached during playback of the video. A dynamic relationship graph is displayed in association with the first time point, where the dynamic relationship graph illustrates one or more relationships among two or more characters in the video, responsive to the first time point being reached. The relationship graph includes two or more nodes and one or more edges. The nodes include a respective node representing each character of the two or more characters in the video, and the edges include a respective edge representing each relationship of the one or more relationships among the characters. It is determined that a second time point is reached in the video. The dynamic relationship graph is updated to represent a change in the one or more relationships between the first time point and the second time point.