Adaptive Image Object Tracking via Posture-Based Algorithm Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object tracking algorithms struggle to maintain accuracy when tracking objects with significant formal variations due to changes in posture or movement direction, leading to potential loss of tracking precision.
Innovation Solution
An image object tracking method and system that determine the moving posture of an object in real-time and select between different tracking algorithms based on the detected posture, using either image feature comparison or motion vector calculation to maintain accurate tracking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If image feature comparison is used for object tracking, then tracking can be performed for objects with consistent appearance, but tracking precision deteriorates when the object has significant formal variations due to posture or movement direction changes
Solution Approach 1:
The patent applies dynamics by making the tracking algorithm selection adaptive and changeable based on real-time posture detection. The system dynamically switches between different tracking algorithms (image feature comparison vs. motion vector calculation) according to the detected moving posture, allowing the tracking mechanism to adapt to varying object appearances and maintain precision across different scenarios
Solution Approach 2:
The patent changes the parameter of tracking algorithm selection based on posture detection results. When the object moves along the first direction, image feature comparison is used; when it moves along the second direction, motion vector calculation is used. This parameter change resolves the contradiction by selecting the appropriate tracking method for each specific scenario
2Measurement precision
If a single tracking algorithm is used for all scenarios, then the system complexity is low, but the tracking precision deteriorates for objects with greater formal variations
Solution Approach 1:
The patent segments the tracking task into different scenarios based on moving posture. By dividing the tracking problem into distinct cases (first direction movement vs. second direction movement), the system can apply specialized algorithms for each segment, improving overall precision while managing complexity through structured division of labor
Solution Approach 2:
The patent creates a universal tracking system that can handle multiple tracking scenarios using a single integrated framework. The posture detection module serves multiple functions by enabling the system to automatically select between different tracking algorithms, making the system multi-functional without requiring separate dedicated systems for each scenario
Data Source
AI summary
A method, a system, and a computer-readable recording medium for image object tracking are provided. A video stream including a previous frame and a current frame is received, wherein the video stream is generated through photographing a space including a moving object by an image capturing device. A moving posture of the moving object determined according to the previous frame is obtained. Whether the moving posture of the moving object is a movement along a first direction or a movement along a second direction is determined. One of different tracking algorithms is selected for performing object tracking according to whether the moving posture of the moving object is the movement along the first direction or the movement along the second direction, to obtain an object feature region associated with the moving object in the current frame. The moving object in the space is tracked according to the object feature region.