Kinematic Sequence Determination via 2D to 3D Coordinate Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current mobile applications for analyzing sports movements rely on multiple hardware sensors and 2D cameras, which are limited in providing accurate 3D analysis of kinematic sequences and detecting movement errors in activities like golf swings.
Innovation Solution
A system utilizing deep learning and computer vision techniques to convert 2D video coordinates to 3D coordinates, allowing for the determination of kinematic sequences and movement errors in golf swings by tracking points on a person's body and golf club, without the need for sensors on the human body.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple hardware sensors (IMU sensors) are attached to several points on a person's body, then measurement precision of movement is improved, but device complexity and ease of operation deteriorate
Solution Approach 1:
The patent extracts the measurement function from physical sensors attached to the body and relocates it to a remote camera system. The camera captures video from a distance, and computational algorithms extract movement data from the video frames, eliminating the need for sensors on the person's body while maintaining measurement capability
Solution Approach 2:
The patent introduces video frames as an intermediary medium between the subject and the measurement system. Instead of direct sensor contact, the camera captures visual information that serves as an intermediary, which is then processed through coordinate extraction and conversion algorithms to derive movement data
2Device complexity
If 2D camera coordinates are used directly, then device complexity is reduced, but measurement precision of 3D movement deteriorates
Solution Approach 1:
The patent applies dimensionality change by converting 2D coordinates extracted from video frames into 3D coordinates through mathematical transformation. The system uses the relationship between camera position, video frame coordinates, and real-world 3D space to calculate three-dimensional positions, enabling accurate 3D kinematic analysis from 2D video input
Solution Approach 2:
The patent replaces the mechanical sensor-based measurement system with a computational vision system. Instead of using physical sensors to directly measure 3D positions, the system uses computer algorithms to convert 2D video coordinates into 3D movement data, substituting mechanical measurement with optical capture and computational processing
3Measurement precision
If sensors are attached to the human body, then measurement precision is improved, but ease of operation and user comfort deteriorate
Solution Approach 1:
The patent removes sensors from the human body and extracts the measurement function to a remote camera system. This eliminates the burden of wearing sensors while maintaining the ability to track movement, significantly improving user convenience and ease of operation
Solution Approach 2:
The patent creates a visual copy of the subject's movement through video capture. Instead of attaching sensors to the body, the system captures a visual representation (video frames) of the movement, which is then processed to extract kinematic data, providing a non-intrusive measurement approach
Data Source
AI summary
Implementations generally relate to determining a kinematic sequence. In some implementations, a method includes obtaining a video of a person performing an action. The method further includes determining from the video a plurality of points associated with the person, where the determining of the plurality of points is performed for each frame of the video. The method further includes converting two-dimensional (2D) coordinates of the plurality of points to three-dimensional (3D) coordinates, where the converting is performed for each frame of the video. The method further includes determining a movement of the plurality of points based at least in part on the 3D coordinates.