Dual-Camera HDR Depth Tracking for Full-Body Exercise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional portable computing devices, such as tablets and mobile phones, face challenges in providing accurate motion tracking and real-time feedback due to their narrow field of view, requiring users to adjust their position and orientation frequently, leading to inconsistencies in capturing full-body movements and compromising user experience, especially in variable lighting conditions.
Innovation Solution
A portable computing device equipped with a dual-camera system and a support member that maintains a substantially vertical orientation, capturing images at different exposure levels to generate high dynamic range (HDR) image data, and utilizing machine learning techniques for improved motion tracking and depth perception, allowing full-body capture without angular adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single camera with narrow field of view is used, then device complexity is reduced, but motion tracking accuracy deteriorates due to inability to capture full-body movements
Solution Approach 1:
The camera system is segmented into multiple cameras (first camera and second camera) with different fields of view. The first camera captures wide-angle full-body movements while the second camera captures close-up facial expressions and on-screen content, allowing each camera to be optimized for its specific function rather than requiring one camera to do everything
Solution Approach 2:
The system transitions from a single-camera 2D view to a multi-camera 3D spatial arrangement. By positioning cameras at different locations and angles, the system captures motion data from multiple dimensions simultaneously, enabling accurate full-body tracking while maintaining device simplicity
2Illumination intensity
If camera exposure is optimized for bright screen content, then on-screen content visibility is improved, but image quality of the user deteriorates due to overexposure in variable lighting conditions
Solution Approach 1:
The image processing is segmented into separate processing pipelines for different exposure levels. The system processes bright screen content and darker user images through different exposure channels, then combines them to produce an HDR image that preserves detail in both bright and dark regions simultaneously
Solution Approach 2:
The system dynamically changes the exposure parameter by capturing images at multiple exposure levels (underexposed, properly exposed, overexposed) and selecting or combining the appropriate exposure data for different regions of the image based on lighting conditions
3Adaptability or versatility
If users must adjust device position and orientation frequently, then motion tracking coverage is improved, but ease of operation deteriorates due to frequent adjustments required
Solution Approach 1:
Instead of requiring users to move the device in space to track different body parts, the system uses multiple cameras positioned at different spatial dimensions to capture the entire body simultaneously from a fixed device position, eliminating the need for frequent adjustments
Solution Approach 2:
The multi-camera system is designed to universally capture various exercise positions and movements (standing, sitting, lying down, arm movements, leg movements) without requiring device repositioning, as each camera contributes to capturing different aspects of the full-body motion
Data Source
AI summary
Systems and methods described herein relate to computer vision-driven full-body motion tracking using a portable computing device. The portable computing device has a first image sensor and a second image sensor to capture images of a full body of the user while the user is performing exercises. The images are captured while the portable computing device is positioned in a substantially vertical orientation. The images are processed to generate high dynamic range (HDR) image data and to determine depth information associated with the user. Motion tracking data is generated in real time, using the HRD image data and the depth information, while the user is performing an exercise. Real-time interactive feedback may be provided during exercises.