Video Object Tracking via Dynamic State Transitions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional object tracking systems in video surveillance struggle with noise and spurious artifacts, leading to inaccurate tracking and loss of object correspondence due to objects appearing, disappearing, occluding, and splitting in video sequences.
Innovation Solution
A method and system for tracking objects by associating status parameters with objects in video frames, using transition rules to update states based on correspondence between frames, and recording intra-frame positions, with features like motion factor and cost functions to determine correspondence and object validity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional object tracking is performed assuming no noise or spurious artifacts, then the tracking process is simple, but tracking accuracy deteriorates when objects appear, disappear, occlude, or split in video sequences
Solution Approach 1:
The patent implements dynamic state transitions for tracked objects, where each object maintains a status parameter that can change between 'tracking' and 'non-tracking' states. This allows the system to adapt to changing conditions (object appearance, disappearance, occlusion, splitting) while maintaining a relatively simple tracking framework. The status parameter dynamically adjusts based on correspondence detection between frames, resolving the contradiction between simplicity and reliability.
2Reliability
If sequential correspondence checking is performed for multiple frames, then tracking reliability improves, but processing time increases
Solution Approach 1:
The patent performs correspondence checking for a predetermined number of sequential frames rather than continuously checking every frame. This partial action approach maintains tracking reliability by verifying correspondence over multiple frames while avoiding the excessive processing time that would result from checking every single frame. The system balances reliability and processing time by limiting the number of sequential checks.
3Measurement precision
If status parameters with transition rules are implemented for each object, then object classification accuracy improves, but system complexity increases
Solution Approach 1:
The patent segments the tracking system by associating individual status parameters with each tracked object rather than using a single global state. Each object's status parameter can be independently updated based on its own correspondence checks, allowing precise classification of each object's state (tracking, non-tracking, occluded, disappeared). This segmentation improves classification accuracy while keeping the overall system manageable by treating each object independently.
Data Source
AI summary
In a digital video surveillance system, a number of processing stages are employed to identify foreground regions representing moving objects in a video sequence. An object tracking stage (5) is also provided in order to identify a correspondence between candidate objects in a current frame and those that have already been identified in one or more previous frames. In this way, it is possible to calculate the path taken by the or each foreground object and to record this path information in a trajectory database. In order to improve tracking performance, the object tracking stage (5) employs a state transitional object management scheme which determines whether or not a particular object is tracked. The path information generated by the object tracking stage (5) can be displayed on a video monitor (15) of the surveillance system and/or made available to high level applications (7).