Posture Estimation Using Subject Size Adjustment in Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing posture detection methods fail to accurately detect the posture of subjects in images when their size deviates from a suitable range, either being too large or too small, leading to inadequate pixel count for detection.
Innovation Solution
A posture estimation system that includes subject detection, size determination, and size adjustment processing to ensure the subject's size falls within a predetermined range, using threshold-based adjustments and camera control for optimal image capture.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the subject size in the captured image is too large, then the subject occupies more pixels, but the subject size deviates from the suitable detection range and posture detection accuracy deteriorates
Solution Approach 1:
The system dynamically adjusts the subject size by selecting between multiple captured images with different subject sizes, rather than using a fixed size adjustment method. This allows the system to adaptively choose the most suitable image for accurate posture detection based on the current subject size conditions
Solution Approach 2:
The system changes the parameter of subject size by selecting images with different subject sizes from multiple captured images. This parameter adjustment ensures the subject falls within the suitable detection range for accurate posture detection while maintaining sufficient pixel quantity
2Measurement precision
If the subject size in the captured image is too small, then the subject fits within the detection range, but insufficient pixels are available for accurate posture detection
Solution Approach 1:
The system dynamically selects between multiple captured images with different subject sizes based on the current subject size conditions. When the subject is too small, it chooses an image where the subject appears larger, thereby obtaining sufficient pixels for accurate posture detection
Solution Approach 2:
The system adjusts the subject size parameter by selecting from multiple images with varying subject sizes. This allows the system to increase the effective subject size and pixel quantity when the original subject is too small, while maintaining it within the detection range
3Measurement precision
If image processing is performed to adjust subject size, then posture detection accuracy improves, but processing time increases
Solution Approach 1:
The system performs preliminary capture of multiple images with different subject sizes before posture detection is needed. This preliminary action allows the system to quickly select the most suitable pre-captured image without performing time-consuming real-time size adjustment processing
Solution Approach 2:
The system uses multiple captured images as copies with different subject sizes. Instead of creating new processed images through complex transformations, it selects from existing image copies that already have the desired subject size characteristics
Data Source
AI summary
A posture estimation apparatus includes a subject detection unit, a size determination unit, a size adjustment processing unit, and a posture detection unit. The subject detection unit detects a subject who is a person appearing in a captured image. The size determination unit determines a size of the subject. The size adjustment processing unit performs processing of adjusting the size of the subject to fall within a predetermined range when the size of the subject is out of the predetermined range. The posture detection unit detects a posture of the subject based on an image in which the size of the subject falls within the predetermined range.