3D Human Model Recovery Using Occlusion Handling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional patient modeling techniques fail to accurately depict a patient's physical characteristics in medical environments where the body is partially blocked by equipment or clothing, leading to unsatisfactory results in applications like radiation therapy and medical imaging.
Innovation Solution
A system utilizing a sensing device, such as a camera or depth sensor, attached to a medical device captures images from different positions to identify body keypoints and generate a 3D human model using machine learning, allowing for the determination of pose and shape parameters within a coordinate system associated with the medical environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional patient modeling techniques are used that assume substantial portions of the body are visible, then the modeling process is simple, but the accuracy of the patient model deteriorates when the body is blocked by medical equipment or clothing
Solution Approach 1:
The system transitions from 2D image analysis to 3D human model reconstruction by capturing images from multiple positions and angles. The sensing device moves around the patient to collect data in three-dimensional space, enabling accurate model generation even when parts of the body are occluded in any single view.
Solution Approach 2:
The system performs preliminary actions by capturing multiple images from different positions before generating the final human model. The sensing device systematically collects data from various angles and positions around the patient, ensuring that occluded regions are captured in at least one unobstructed view before model reconstruction begins.
2Measurement precision
If multiple images are captured from different positions to handle occlusions, then the accuracy of the human model improves, but the complexity of the system increases
Solution Approach 1:
The sensing device is designed with multi-functionality, serving both as an image capture device and a positioning system. The device can capture images from multiple positions and simultaneously track its own location within the coordinate system, eliminating the need for separate positioning equipment and reducing overall system complexity.
Solution Approach 2:
A coordinate system serves as an intermediary that links the sensing device positions with the captured images and the final human model. By establishing a unified coordinate framework, the system seamlessly integrates multi-position image data without requiring complex registration or alignment algorithms.
3Manufacturing precision
If the sensing device is attached to a medical device to capture images during treatment, then the patient positioning accuracy improves, but the device complexity and operational complexity increase
Solution Approach 1:
The sensing device is merged with the medical device, combining patient imaging and positioning functions into a single integrated system. This merger allows the medical device to simultaneously perform its primary medical function and capture images for human model generation, eliminating the need for separate equipment and simplifying operation.
Solution Approach 2:
The integrated system performs self-service by automatically capturing images and generating human models during the normal operation of the medical device. The system autonomously tracks the medical device position and uses this information to reconstruct the patient model without requiring additional manual intervention or complex operational procedures.
Data Source
AI summary
A human model such as a 3D human mesh may be generated for a person in a medical environment based on one or more images of the person. The images may be captured using a sensing device that may be attached to an existing medical device such as a medical scanner in the medical environment. Such an arrangement may ensure that unblocked views of the person (e.g., body keypoints of the person) may be obtained and used to generate the human model. The position of the medical device in the medical environment may be determined and used to facilitate the human model construction such that the pose and body shape of the person in the medical environment may be accurately represented by the human model.