Body Hand Correlation via Wrist Elbow Key Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In complex scenes with many people, existing methods for body and hand correlation are prone to errors due to occlusions and staggered arms, leading to inaccurate judgment and high costs, as they rely on human key point detection combined with sensors.
Innovation Solution
A method that determines correlation probabilities between body and hand detection boxes and wrist key points, using features and positional relationships to optimize the accuracy of hand-to-body correlation, reducing dependence on body detection box accuracy and simplifying the prediction of hand positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If human key point detection combined with sensors is used to judge body-hand correlation, then judgment can be made, but judgment accuracy deteriorates and computational cost increases
Solution Approach 1:
The patent extracts and utilizes only the necessary detection boxes and key points from the complex scene, filtering out irrelevant information. By focusing on the body detection box, hand detection box, and their corresponding key points (wrist, elbow, shoulder), the method simplifies the computational process while maintaining accurate correlation judgment without requiring additional sensors or complex processing
Solution Approach 2:
The patent segments the body and hand into distinct detection boxes and identifies specific key points within each segment. This segmentation allows for independent processing and correlation calculation between the body detection box and hand detection box, reducing the computational complexity by breaking down the overall task into manageable parts
2Measurement precision
If body detection box accuracy is heavily dependent, then correlation can be determined, but the system becomes sensitive to detection errors in complex scenes
Solution Approach 1:
The patent introduces key points (wrist, elbow, shoulder) as intermediary elements that mediate between the body detection box and hand detection box. These key points serve as reference markers that can be reliably detected even in complex scenes, allowing the system to determine correlation based on the spatial relationship between key points rather than solely relying on the accuracy of the bounding boxes
Solution Approach 2:
The patent changes the parameter used for correlation determination from the bounding box coordinates to the key point coordinates. By using key points (wrist, elbow, shoulder) as the reference parameters, the system becomes more robust to detection errors in complex scenes, as key points can be reliably identified even when body contours are occluded or ambiguous
Data Source
AI summary
Body and hand correlation method, apparatus, device, storage medium and computer program are provided. Method includes: in an image of which an image content includes a body to be correlated and a hand to be correlated, a first correlation probability between a body detection box of the body and a hand detection box of the hand is determined; a second correlation probability between the body and a wrist key point in a key point is determined based on the key point in the body detection box, the key point including the wrist key point and elbow key point belonging to the same arm; a third correlation probability between the hand detection box and the wrist key point is determined based on the wrist key point and elbow key point; and a correlation degree between the body and the hand is determined based on the first, second, and third correlation probabilities.