Video Indicator Overlay for Real-Time Object Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Viewers find it difficult to track fast-moving objects and participants in sporting and e-sports events due to the speed and small size of the objects relative to the field and players, making it challenging to follow the action during live broadcasts.
Innovation Solution
A system and method that generate an indicator for display in the video to aid users in tracking key instruments in events by identifying the type of sport and applying indicators such as halos, pointers, or animations to the key objects or participants, using image recognition and machine learning models, which can be applied in real-time or post-capture, and can be transmitted to various devices for broadcast.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If no indicator is applied to objects in the video, then the video stream remains unprocessed and simple to transmit, but viewers cannot effectively track fast-moving objects and participants
Solution Approach 1:
The system performs preliminary actions by pre-identifying key objects and participants in the video stream before transmission. Metadata is generated in advance that contains information about object locations, types, and tracking parameters, which are then overlaid on the video frames. This preliminary processing enables viewers to track objects effectively without requiring complex real-time processing at the receiving end.
Solution Approach 2:
The patent introduces an intermediary metadata layer between the raw video stream and the viewer. This metadata acts as a mediator that carries tracking information about key objects and participants, enabling the viewer to follow the action without directly processing the complex visual data. The metadata includes object identifiers, positions, and other relevant information that simplifies the tracking task for the end user.
2Loss of information
If indicators are applied to all frames of the video, then continuous tracking information is provided to viewers, but the amount of data to transmit and process increases significantly
Solution Approach 1:
The system applies indicators periodically rather than continuously to every frame. By selecting specific frames (e.g., every nth frame) to receive indicator overlays, the system maintains sufficient tracking information for viewers while dramatically reducing the total data volume. This periodic application balances the need for tracking continuity with the constraints of data transmission and processing capacity.
3Speed
If real-time indicator application is implemented, then viewers receive immediate tracking information during live events, but significant processing power is required
Solution Approach 1:
The system performs preliminary identification and tracking setup before the live event begins or during minimal buffering periods. Key objects and participants are identified in advance, and tracking parameters are pre-calculated. During the actual live broadcast, the system only needs to apply pre-determined indicators to appropriate frames, significantly reducing the real-time processing power requirements while maintaining immediate tracking information for viewers.
Data Source
AI summary
Systems and methods are provided for identifying a key instrument in an event. One example method includes receiving a capture of the event and identifying, at a first computing device, the event. The key instrument is identified at the first computing device. An indicator to apply to and/or around the identified key instrument is generated for display.