Pose Identification Using Multi-Region Body Pattern Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pose identification techniques in images suffer from decreased accuracy when image resolution is low, as they rely solely on head detection, leading to inefficiencies in identifying multiple individuals in low-resolution images.
Innovation Solution
A pose identifying apparatus and method that acquires information about the position and type of detection points for multiple body region points in images, classifying these points to identify poses by extracting a basic pattern including a reference body region point and multiple base body region points, which are different from the reference type, thereby improving identification accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If head-based identification is used, then the identification process is simple, but the accuracy decreases in low-resolution images
Solution Approach 1:
The identification system segments the human body into multiple detectable regions (head, body, limbs) rather than relying solely on head detection. Each body region is independently detected and classified, allowing the system to maintain accuracy even when image resolution is low, as multiple larger body regions provide more detectable features
Solution Approach 2:
The system transitions from one-dimensional head-only detection to multi-dimensional body region detection by adding spatial dimensions (body, limbs, etc.). This dimensional expansion allows the system to gather more information from the same image, improving identification accuracy without significantly increasing processing complexity
2Speed
If only head detection is used, then the processing is fast, but the reliability of identification decreases
Solution Approach 1:
The system merges multiple detection results (head, body, limbs) into a unified identification process. By combining information from multiple body regions, the system increases identification reliability through cross-validation, while maintaining processing efficiency by using parallel detection algorithms
3Measurement precision
If basic pattern extraction with multiple body region points is used, then identification accuracy improves, but the device complexity increases
Solution Approach 1:
The system performs preliminary extraction of basic body patterns (head, body, limbs) before detailed pose identification. These pre-extracted patterns serve as standardized templates that simplify subsequent classification and matching processes, reducing overall system complexity while maintaining high identification accuracy
Data Source
AI summary
A basic pattern extracting unit (15) extracts a “basic pattern” for each human from detection points acquired by an acquiring unit (11). The “basic pattern” includes a “reference body region point” corresponding to a “reference body region type”, and base body region points corresponding to base body region types that are different from the reference body region type and that are different from each other. For example, the “basic pattern” includes at least one of the following two combinations. A first combination is a combination of the reference body region point corresponding to a neck as the reference body region type and two base body region points respectively corresponding to a left shoulder and a left ear as the base body region type. A second combination is a combination of body region points corresponding to a neck, a right shoulder and a right ear.