Object Tracking via Dynamic Region Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing artificial reality systems face challenges in efficiently tracking objects in real-time due to the high computational resources and large data quantities required, especially in virtual reality, augmented reality, and mixed reality applications, where precise location determination of objects is necessary for interactive displays.
Innovation Solution
A system comprising one or more imaging devices and an object tracking unit that identifies objects in a search region, determines a smaller tracking region, and scans it over time to determine precise locations, improving speed and efficiency by generating smaller frames of image data for processing, and can generate a model of the environment based on object locations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If the system scans the entire search region to track objects, then the coverage area is comprehensive, but the processing time and computational resources increase significantly
Solution Approach 1:
The patent divides the large search region into multiple smaller tracking regions. After initial object detection in the search region, the system identifies bounding boxes and creates smaller tracking regions around detected objects. This segmentation allows the system to process only relevant areas in subsequent frames, reducing computational load while maintaining comprehensive coverage through periodic full search region scans.
Solution Approach 2:
The system performs preliminary object detection in the full search region first, then uses the detected object positions to define tracking regions for subsequent frames. This preliminary action allows the system to prepare tracking regions in advance based on object locations, so that future processing can focus only on these pre-defined smaller regions rather than re-scanning the entire search area.
2Measurement precision
If the system processes large search region image data, then the object detection is thorough, but the computational resources and data quantities required increase
Solution Approach 1:
The patent extracts only the necessary portions of the search region image data by creating tracking regions around detected objects. Instead of processing the entire large search region image data in every frame, the system extracts and processes only the smaller tracking region image data that contains the objects of interest, significantly reducing the quantity of data that needs to be processed while maintaining detection accuracy.
3Productivity
If the system uses smaller tracking regions for object tracking, then the processing speed increases, but the risk of missing objects or losing track increases
Solution Approach 1:
The system implements feedback mechanisms where it periodically re-scans the full search region to detect new objects that may have entered the scene. It also monitors object positions and adjusts tracking regions dynamically based on object movement. This feedback ensures that while using smaller tracking regions for speed, the system maintains reliability by continuously verifying object presence and updating tracking parameters based on detected object behavior.
Data Source
AI summary
Embodiments relate to tracking and determining a location of an object in an environment surrounding a user. A system includes one or more imaging devices and an object tracking unit. The system identifies an object in a search region, determines a tracking region that is smaller than the search region corresponding to the object, and scans the tracking region to determine a location associated with the object. The system may generate a ranking of objects, determine locations associated with the objects, and generate a model of the search region based on the locations associated with the objects.