Object Tracking via Feature-Based Cost Function
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video analytics solutions for object tracking in videos and images face high computational complexity, making them unsuitable for embedded platforms, despite their potential for accurate tracking.
Innovation Solution
An object tracking system that computes a cost function based on visual dissimilarity, physical distance, and expected location between objects in successive frames, using a processor and memory unit to store and select pairs of objects for accurate tracking, optimizing computational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If online learning based strategies with discriminative classifier models (such as SVM) are used for object tracking, then tracking accuracy is improved, but computational complexity increases making it unsuitable for embedded platforms
Solution Approach 1:
The patent segments the tracking problem into distinct components: object detection, feature extraction, and matching. By dividing the complex discriminative classification task into smaller, more manageable sub-tasks, the system achieves comparable tracking accuracy with reduced computational complexity suitable for embedded platforms.
Solution Approach 2:
The patent applies local quality by focusing computational resources on extracting and comparing specific features (such as color histograms, shape descriptors) at object locations rather than processing entire frames through complex classifiers. This localized feature-based approach maintains tracking precision while significantly reducing overall computational burden.
2Measurement precision
If complex video analytics algorithms are implemented, then object tracking precision is improved, but power consumption increases making it unsuitable for embedded platforms
Solution Approach 1:
The patent implements partial action by performing feature extraction and matching only on detected object regions rather than processing entire video frames through power-intensive algorithms. This selective processing approach maintains tracking precision while significantly reducing power consumption for embedded platform operation.
3Productivity
If traditional object tracking methods (MeanShift, Optical Flow) are used, then computational efficiency is improved, but tracking accuracy deteriorates compared to online learning based strategies
Solution Approach 1:
The patent changes the parameters used for tracking from traditional motion-based features (MeanShift, Optical Flow) to appearance-based features such as color histograms, texture descriptors, and shape features. This parameter transformation enables the system to achieve higher tracking accuracy while maintaining computational efficiency suitable for embedded platforms.
Data Source
AI summary
An embodiment of the present invention discloses an object tracking system for tracking objects across a first frame and a second frame of a video. The object tracking system comprises of a processor, a non-transitory storage element coupled to the processor and encoded instructions stored in the non-transitory storage element. The encoded instructions when implemented by the processor, configure the object tracking system to detect one or more objects in the first frame of the video, and one or more candidate objects in the second frame of the video.