Markerless Motion Capture via Silhouette Tracking and Joint Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current motion capturing technologies require expensive equipment and are sensitive to environmental changes, leading to noise in captured images, necessitating a cost-effective and noise-reduced solution.
Innovation Solution
A markerless motion capturing apparatus that calculates body information and determines motion trajectories from images, tracks silhouettes, and detects joints using a combination of units such as motion tracking information determination, silhouette tracking, and joint detection, potentially with multiple cameras for 2D and 3D space analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If markers or sensors are attached to the performer's body for motion capture, then motion tracking accuracy is improved, but equipment cost and system complexity increase
Solution Approach 1:
The patent extracts and removes the markers and sensors from the motion capture system, achieving markerless motion capture. The system processes natural video images directly to extract motion information, eliminating the need for special equipment attached to the performer while maintaining motion tracking capability through computer vision algorithms
Solution Approach 2:
The patent uses video camera images as a copy or representation of the performer's actual motion. Instead of directly measuring motion with sensors, the system captures optical copies (video images) and processes them to extract motion trajectories, joint positions, and body pose information
2Speed
If sensors are used to capture performer motion, then real-time tracking is achieved, but environmental sensitivity causes noise in captured data
Solution Approach 1:
The patent replaces the mechanical sensor-based detection system with an optical vision-based system. Video cameras capture motion optically, and computer vision algorithms process the images to extract motion information, substituting physical sensors with optical-field-based measurement that is less sensitive to environmental changes
3Measurement precision
If multiple cameras are used for 3D motion capture, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent transitions from 2D video images to 3D motion capture by introducing depth information extraction. Multiple cameras capture images from different viewpoints, and the system processes these multi-view images to reconstruct three-dimensional joint positions and body poses, adding the depth dimension to the measurement space
Data Source
AI summary
A markerless motion capturing apparatus and method is provided. The markerless motion capturing apparatus may track a pose and a motion of a performer from an image, inputted from a camera, without using a marker or a sensor, and thereby may extend an application of the markerless motion capturing apparatus and selection of a location.