3D Body Modeling Using Cylindrical Segmentation for Articulated Motion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 3D body modeling techniques face challenges in generating accurate models from range sensor data, especially in the presence of motion, as they struggle to handle articulated body parts and require multiple cameras or complex calibration processes.
Innovation Solution
The method involves capturing 3D point clouds using a single or multiple 3D cameras, transforming and segmenting them into cylindrical representations, and using a 3D part-based volumetric model to account for localized articulated motion, enabling accurate registration and generation of 3D body models even with motion, through techniques like iterative closest point (ICP) and cylindrical image representations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple 3D cameras are used to capture 3D point clouds, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the body into multiple cylindrical parts (torso, head, limbs) and processes each part separately through coordinate transformation. This segmentation allows accurate handling of articulated motion by treating each body part independently, resolving the contradiction between using multiple cameras and system complexity by achieving multi-view accuracy through single-camera multi-part processing
Solution Approach 2:
The patent transforms 3D point cloud data into cylindrical coordinate systems for each body part, adding a dimensional transformation layer. This allows the system to handle articulated motion more effectively by representing body parts in a coordinate system that naturally accommodates rotational joints, achieving multi-camera-like accuracy through dimensional transformation rather than physical multi-camera setup
2Manufacturing precision
If multiple 3D cameras are used to capture 3D point clouds, then manufacturing precision is improved, but ease of manufacture deteriorates
Solution Approach 1:
The patent creates a universal cylindrical representation framework that can process point cloud data from either single or multiple cameras. The same coordinate transformation and cylindrical modeling algorithms work regardless of the number of cameras, making the system universally applicable and easier to implement while maintaining high accuracy through the robust cylindrical part-based approach
Solution Approach 2:
The patent creates virtual copies of body parts in a standardized cylindrical coordinate system. By transforming real-world point cloud data into standardized cylindrical representations, the system achieves consistent accuracy without requiring complex multi-camera calibration, simplifying implementation while maintaining manufacturing precision
3Measurement precision
If transforms are determined for each captured 3D point cloud to transform to reference point cloud coordinates, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent performs preliminary segmentation of the reference point cloud into cylindrical body parts before processing transformed point clouds. By pre-defining the cylindrical framework and transformation targets, the system reduces the computational complexity of subsequent registration operations, maintaining measurement precision while reducing the time lost in processing
Solution Approach 2:
The patent segments both reference and transformed point clouds into corresponding body parts (torso, head, limbs) and performs registration on segmented parts rather than entire point clouds. This segmentation reduces the number of points that need to be transformed and matched, significantly reducing processing time while maintaining or improving registration accuracy through localized part-based matching
4Ease of operation
If the body is segmented into body parts corresponding to cylindrical representations, then ease of operation is improved, but device complexity increases
Solution Approach 1:
The patent uses dynamic cylindrical representations that can accommodate articulated motion of body parts. Each cylindrical body part can independently transform relative to others, allowing the model to dynamically adapt to different poses and movements. This dynamic approach simplifies operation with moving subjects while the underlying cylindrical-part complexity is managed through systematic coordinate transformations
Data Source
AI summary
The present disclosure describes systems and techniques relating to generating three dimensional (3D) models from range sensor data. According to an aspect, 3D point clouds are captured using a 3D camera, where each of the 3D point clouds corresponds to a different relative position of the 3D camera with respect to a body. One of the 3D point clouds can be set as a reference point cloud, and transforms can be determined for coordinates of the other captured 3D point clouds to transform these to coordinates of the reference point cloud. The body represented in the reference point cloud can be segmented into body parts corresponding to elements of a 3D part-based volumetric model including cylindrical representations, and a segmented representation of the physical object of interest can be generated in accordance with the 3D part-based volumetric model, while taking localized articulated motion into account.