Skeleton Keypoint Quality Evaluation for Multi-Camera Template Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The accuracy of detecting human bodies with desired poses and movements from images decreases without registering an image of certain quality as a template, and there is a need to improve the workability of preparing such a template image.
Innovation Solution
An image processing apparatus that includes a skeleton structure detection unit to detect keypoints of a human body in multiple images from multiple cameras, a determination unit to identify the same human body across images, a quality value computation unit to compute a quality value for each detected keypoint, and an output unit to provide information on the location of human bodies with quality values meeting a threshold or to output partial images from those locations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an image of certain quality is registered as a template image to improve detection accuracy, then the accuracy of detecting human bodies with desired poses and movements is improved, but the workability of preparing the template image deteriorates
Solution Approach 1:
The system automatically evaluates image quality and selects template images without requiring manual intervention. The quality evaluation unit computes quality values based on skeleton structure detection, and the template image selection unit automatically chooses images meeting the quality threshold, eliminating the need for manual template preparation while ensuring high detection accuracy
Solution Approach 2:
The system changes the parameter of image quality assessment by introducing a quantitative quality value computation based on skeleton structure analysis. This objective parameter-based selection replaces subjective manual judgment, automatically identifying images with clear skeleton structures suitable for template registration, thus improving both detection accuracy and preparation efficiency
2Measurement precision
If multiple cameras are used to capture images for template preparation, then the quality of detected keypoints is improved, but the complexity of the system increases
Solution Approach 1:
The image processing apparatus performs multiple functions using the same multi-camera system: it captures images for both quality evaluation and template preparation, detects skeleton structures for quality assessment, and selects template images automatically. This multi-functional approach leverages the existing camera infrastructure without adding separate specialized equipment, maintaining system simplicity while improving keypoint detection quality
Solution Approach 2:
The quality evaluation unit acts as an intermediary between the multi-camera system and the template image selection process. It computes quality values based on skeleton structure detection from multiple camera images and filters images based on quality thresholds, mediating the complex multi-camera input into a simplified template selection output without requiring direct complex interaction between cameras
Data Source
AI summary
The present invention provides an image processing apparatus (10) including: a skeleton structure detection unit (11) that performs processing of detecting a keypoint of a human body included in each of a plurality of images generated by a plurality of cameras capturing a same place; a determination unit (12) that determines a same human body included in the plurality of images generated by the plurality of cameras; a quality value computation unit (13) that computes, for each human body, a quality value of the keypoint detected from the plurality of images generated by the plurality of cameras; and an output unit (14) that outputs information indicating a place where a human body with the quality value equal to or more than a threshold value is captured, or a partial image acquired by cutting the place out of the image.