Mobile Camera User Analytics with Pose and Depth Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital coaching and training applications are passive and offline, requiring manual analysis of video recordings, which limits real-time feedback and accuracy in determining user location and movement information in three-dimensional environments, especially with limited computational resources of mobile devices.
Innovation Solution
A computer-implemented method using a mobile device with a camera, employing machine learning and machine vision algorithms for real-time determination of user location and movement, including pose estimation, depth calculation, and noise filtering, to provide accurate and efficient analytics in a three-dimensional environment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual analysis of video recordings is used, then accuracy in determining user location and movement information can be improved, but productivity and real-time feedback capability deteriorate
Solution Approach 1:
The patent replaces manual mechanical analysis with automated computer vision algorithms and machine learning models that process video data in real-time. The system uses pose estimation algorithms, depth calculation, and noise filtering to automatically determine user location and movement information, eliminating the need for manual frame-by-frame analysis while maintaining high accuracy and enabling real-time feedback.
2Measurement precision
If complex algorithms are used for accurate pose estimation and depth calculation, then measurement precision improves, but device complexity and computational resource requirements worsen
Solution Approach 1:
The patent divides the complex analysis task into multiple independent algorithmic modules: user identification, pose estimation, depth calculation, and noise filtering. Each module processes specific aspects of the data independently, allowing for optimized computation and reducing the overall complexity burden on mobile devices while maintaining high measurement precision through specialized processing for each function.
Solution Approach 2:
The system implements selective processing by focusing computational resources on key parameters that most impact measurement accuracy, such as critical pose points and essential depth information. The noise filtering algorithm selectively processes data points based on their relevance and quality, performing partial analysis on less critical data while maintaining high precision for essential measurements, thereby reducing overall computational requirements.
Data Source
AI summary
Methods and systems for performing a location determination in a three-dimensional environment using a computing device with a camera are disclosed. The methods and systems perform the steps of receiving an image of a user from the camera of the computing device. Identifying the user from the image using a first algorithm, where the first algorithm is a machine learning algorithm. Determining a pose information associated with the user using a second algorithm, where the second algorithm is a machine vision algorithm. Determining a depth information associated with the user based on the pose information and an input parameter (e.g., height) of the user using a search process (e.g., binary search). Finally, determining the location of the user in the environment based on the pose information and the depth information.