Pose Estimation Confidence Filtering Using Keypoints and Bounding Boxes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional bottom-up pose estimation techniques are prone to errors due to misdetection and incorrect connections of keypoints, leading to unnatural and erroneous pose estimation results.
Innovation Solution
An image processing apparatus that detects keypoints and bounding boxes for subjects, and adjusts confidence levels based on their positional relationships to reduce errors in pose estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If bottom-up pose estimation technique is used, then computational cost is reduced, but detection precision and reliability deteriorate due to misdetection and incorrect connections of keypoints
Solution Approach 1:
The system performs feedback by comparing the positional relationship between detected keypoints and the extracted bounding box, then uses this comparison to adjust confidence levels. When keypoints fall outside the bounding box or show inconsistent spatial relationships, the system reduces their confidence levels, creating a closed-loop verification mechanism that improves detection reliability without requiring additional computational resources for re-detection.
Solution Approach 2:
The system dynamically changes the confidence level parameter of keypoints based on their positional relationship with the bounding box. By adjusting this parameter according to spatial consistency checks, the system filters out misdetections and incorrect keypoint connections, thereby improving measurement precision while maintaining the efficiency of the bottom-up approach.
2Device complexity
If bottom-up pose estimation technique is used, then device complexity is reduced, but reliability deteriorates due to erroneous pose estimation results
Solution Approach 1:
The bounding box serves as an intermediary element that mediates between the detected keypoints and the final pose estimation. By introducing this intermediate verification step, the system checks whether keypoints are spatially consistent with the bounding box before accepting them for pose estimation, thereby improving reliability without adding complex device architecture.
3Reliability
If confidence level adjustment based on positional relationship is implemented, then reliability is improved, but device complexity increases
Solution Approach 1:
The system performs self-service by automatically adjusting confidence levels based on intrinsic spatial relationship checks between keypoints and the bounding box. This self-verification mechanism improves reliability without requiring external verification systems or complex additional processing, as the system uses its own detection outputs to validate itself.
Data Source
AI summary
An image processing apparatus is configured to reduce erroneous results in pose estimation for a subject. The image processing apparatus detects a plurality of keypoints for a subject in an image and determines whether the keypoints belong to the same subject. The image processing apparatus extracts a bounding box that encloses a body part of a subject in the image and indicates a detection range of the subject, and determines whether the bounding box corresponds to the same subject as the keypoints. According to at least a positional relationship between the keypoints determined to belong to the same subject and the bounding box determined to correspond to the same subject as the keypoints, the confidence level of at least one of the bounding box and the keypoints is reduced, or the confidence level of pose estimation for the subject using the keypoints is reduced.