Automated Play Card Generation from Image Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In team sports, creating and managing play templates for practice sessions is resource-intensive and time-consuming for coaches, especially as the competitive level increases, requiring frequent changes in play templates.
Innovation Solution
The use of image data to automatically generate play cards by deriving play formation and individual player movement models, which are then combined into a play card data file and transmitted to wearable player electronic devices for display.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If coaches manually create play templates for each practice session, then detailed play information can be provided to players, but significant time and resources are consumed in the creation process
Solution Approach 1:
The system captures actual play execution data from practice sessions using tracking technology and creates digital templates that automatically replicate the observed plays. Instead of manually creating templates from scratch, the system copies real player movements and formations to generate accurate play templates, eliminating the time-consuming manual creation process while preserving detailed play information.
Solution Approach 2:
The patent replaces the manual mechanical process of coaches creating play templates with an automated computer vision and machine learning system. Image data from practice sessions is processed algorithmically to extract play formations and movements, substituting human manual work with automated computational processes that generate play templates rapidly and accurately.
2Reliability
If coaches manually create play templates before each practice, then plays can be executed with proper guidance, but the opportunity cost of not devoting that time to more valuable uses increases
Solution Approach 1:
The system enables play templates to be automatically generated from captured practice data without requiring coach intervention for creation. The automated system serves itself by capturing, processing, and generating play templates independently, allowing coaches to focus on higher-value activities while the system handles the template generation autonomously.
Solution Approach 2:
The system performs preliminary capture of play data during practice sessions and automatically generates templates in real-time or near-real-time. By preparing play information automatically as plays are executed, the system eliminates the need for pre-practice template creation while ensuring play guidance is available when needed.
3Adaptability or versatility
If the competitive level of sport increases, then more play templates need to be created to maintain effectiveness, but the time and resources required to create these templates increase proportionally
Solution Approach 1:
The system creates a universal template generation platform that can handle multiple play types and competitive levels through a single automated system. The machine learning models are trained to recognize various formations and plays across different sports and competitive levels, allowing one system to generate diverse play templates without requiring separate manual creation processes for each play type.
Solution Approach 2:
The system adapts to different competitive levels by adjusting parameters such as play complexity, formation variations, and movement patterns. The automated generation process can modify template parameters to match the required competitive level, generating appropriate numbers of play templates for each practice session based on the specific needs of the team and competition level without proportional increases in creation time.
Data Source
AI summary
A computing device has non-transitory computer-executable instructions that, when executed by programmable processing circuitry of the computing device, cause the programmable processing circuitry of the computing device to: derive a play formation model from player positional formation data that is extracted from image data of a play prior to start of the play; derive a plurality of individual player movement models from player movement data that is extracted from image data of the play being run; combine the derived play formation model and the derived plurality of individual player movement models to create a play card data file; and transmit the play card data file to a plurality of wearable player electronic devices for display of the play card data file.