Part-Based Object Tracking for Real-Time Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for detecting and tracking objects from images suffer from high processing loads, making real-time detection and tracking at 30 frames per second difficult.
Innovation Solution
An information processing device that tracks objects in images using a control unit to perform tracking in units of specific parts of the object, utilizing a tracking region, and employs time delay compensation and importance level determination to reduce processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If object detection is performed using existing techniques (e.g., CNN-based part point detection), then detection accuracy is improved, but processing load increases making real-time tracking difficult
Solution Approach 1:
The patent divides the object tracking task into multiple tracking regions corresponding to specific parts of the object (e.g., head, body, limbs). Each tracking region is processed independently with dedicated tracking algorithms, allowing parallel processing and reducing overall computational load while maintaining detection accuracy through multi-region coverage
2Reliability
If tracking is performed on the entire object, then tracking completeness is improved, but processing load increases
Solution Approach 1:
The object is divided into multiple tracking regions (head, body, limbs) that are tracked independently. This segmentation allows the system to maintain complete object tracking by combining results from multiple regions while reducing processing load through distributed computation and focused attention on critical regions
Solution Approach 2:
Different tracking algorithms and processing intensities are applied to different tracking regions based on their importance and motion characteristics. Critical regions like the head receive more processing resources, while less critical regions use simpler tracking, optimizing the balance between completeness and efficiency
Data Source
AI summary
An information processing device that includes a control unit configured to track an object in an image using images input in time series, using a tracking result obtained by performing tracking in units of a tracking region corresponding to a specific part of the object.