Deriving Subsidiary Competitions from Video Matchups
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing competitions, such as sports leagues or tournaments, do not effectively utilize shorter-duration events like dunks, isolations, or blocks as standalone competitions, limiting the ability to derive meaningful subsidiary competitions from these events.
Innovation Solution
The method involves deriving competitions from event video, context, and statistics of an original competition by specifying relevant event types and rules, using spatiotemporal pattern recognition and machine learning to identify and rank matchup events, and presenting them as video clips with graphics and statistics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If shorter-duration events are viewed only as components of larger tournaments, then the structure of existing competitions is maintained, but the ability to derive meaningful subsidiary competitions is limited
Solution Approach 1:
The patent segments the original competition into multiple independent subsidiary competitions based on different event types (e.g., dunks, isolations, blocks). Each event type becomes a separate competition with its own ranking system, allowing the system to derive meaningful subsidiary competitions while maintaining manageable complexity through modular organization.
Solution Approach 2:
The patent adds a new dimension to competition analysis by creating temporal and thematic subdivisions of events. Instead of viewing events only within the context of the overall tournament, the system creates additional competition layers based on event type and timing, transforming the single-dimension tournament structure into a multi-dimensional competition framework.
2Loss of information
If individual player performances in specific event types are highlighted, then entertainment value and detailed analysis are improved, but the complexity of identifying and ranking matchup events increases
Solution Approach 1:
The patent creates simplified copies or representations of the complex matchup events. By defining standardized event types (dunks, isolations, blocks) with specific detection criteria, the system creates manageable event models that capture essential player performance information without requiring complex analysis of every game moment, thus reducing identification difficulty while preserving key performance data.
Solution Approach 2:
The patent transforms the detection problem by changing the parameters used to identify and measure events. Instead of analyzing continuous game footage for any possible interaction, the system defines discrete event types with specific measurable parameters (e.g., dunk completion, isolation duration, block timing), making detection and ranking more straightforward while capturing detailed player performance information.
Data Source
AI summary
Video feeds may be analyzed to identify matchups between participants. The matchups may be used to derive a competition, context, statistics, and the like. A result of the derived competition may depend, at least in part, on an ordering and outcomes of a set of matchup events. The derived competition may be presented as a collection of video clips corresponding to the set of matchup events, related commentary, and/or related statistics.


