Real-Time Object Selection in Live Video Relay Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional relay broadcasting systems do not allow users to actively select and view specific objects within broadcasted pictures, limiting viewer engagement and screen selection to only edited information provided by the broadcasting station.
Innovation Solution
A method and system that extracts objects from multiple camera-generated picture information, assigns identifiers, classifies and tracks object positions, and displays position map information in real-time, enabling users to select and view desired objects and their corresponding pictures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional relay broadcasting is used to collect and broadcast picture information from multiple cameras, then the broadcasting station can provide edited picture information to viewers, but the viewers cannot actively select specific objects or pictures and are limited to passively viewing unilaterally broadcasted content
Solution Approach 1:
The system segments the broadcast content by extracting and tracking specific objects (players, balls, etc.) from the video feed, creating separate object streams that can be independently selected by users. This allows users to choose specific objects of interest without requiring complex manual editing of entire video streams.
Solution Approach 2:
The patent introduces an intermediary system comprising object extractors, position trackers, and map generators that act as a mediator between the camera feeds and the user interface. This intermediary layer automatically processes video data to create selectable object representations, reducing the need for complex manual editing systems while enabling user selection capability.
2Adaptability or versatility
If the system extracts and tracks multiple objects from picture information to enable user selection, then users can actively select desired objects and pictures, but the processing complexity and computational requirements increase
Solution Approach 1:
The system implements self-service by enabling objects to automatically identify and track themselves through assigned identifiers. Each extracted object maintains its own position data and can be independently selected by users without requiring complex system-wide processing for each selection request.
Solution Approach 2:
The patent transforms video data into different parameter representations including object positions, velocities, and identified characteristics. By changing the data parameters from raw video pixels to structured object attributes, the system enables efficient tracking and selection while reducing processing complexity for user interactions.
3Ease of operation
If real-time position tracking and mapping is implemented to display object locations, then users can smoothly select objects and view corresponding pictures, but the computational load and processing time increase
Solution Approach 1:
The system performs preliminary actions by pre-extracting objects, assigning identifiers, and generating position maps in advance before user selection is needed. This preprocessing creates ready-to-use object data structures that can be quickly accessed and displayed when users interact with the interface, reducing real-time processing delays.
Data Source
AI summary
Provided is a method and system for displaying a user selectable picture that enables a user to select a predetermined object from a picture of a camera relayed in real time, and to actively select and thereby view a picture captured to include the selected object.