Video Pose Analysis for Movement Error Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Evaluating and improving an individual's movement characteristics, such as posture and mobility, often requires subjective analysis by experts, leading to potential errors and suboptimal exercise recommendations.

Innovation Solution

A system that uses video and sensor data to evaluate user movements, classify errors, and recommend exercises to improve movement characteristics without expert intervention, utilizing machine learning to analyze pose data and provide personalized feedback.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If expert analysis is used to evaluate movement characteristics, then measurement precision is improved, but device complexity and loss of time increase

Engineering Contradiction:
Improvemovement evaluation accuracyVSAvoidexpert intervention requirement
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system enables users to independently evaluate their own movement characteristics through automated video analysis and machine learning algorithms, eliminating the need for expert intervention while maintaining evaluation accuracy through objective computational assessment

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical system of expert human analysis with an automated computational system that uses video capture, pose estimation algorithms, and machine learning models to objectively evaluate movement characteristics, thereby reducing device complexity and time loss

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If expert analysis is used to evaluate movement characteristics, then measurement precision is improved, but loss of time increases

Engineering Contradiction:
Improvemovement evaluation accuracyVSAvoidevaluation time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system provides continuous automated evaluation of movement characteristics during exercise performance, with real-time video analysis and immediate feedback, eliminating the intermittent and time-consuming nature of expert review while maintaining measurement precision through sustained computational assessment

Inventive Principle:
Principle #20Continuity of useful action

3Device complexity

If automated video analysis is used to evaluate movement, then device complexity is reduced, but measurement precision may worsen

Engineering Contradiction:
Improvesystem simplicityVSAvoidmovement evaluation accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The system transforms complex movement evaluation into measurable parameters through pose estimation algorithms that extract key body joint positions and angles from video data, enabling automated analysis to achieve measurement precision comparable to expert assessment while maintaining device simplicity

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent replaces expert mechanical analysis with computational algorithms that use machine learning models trained on expert-labeled data, allowing automated systems to achieve high measurement precision through pattern recognition and classification without requiring physical expert intervention

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4295328B1System for correcting user movement
Publication Date: 2026.02.04 AMAZON TECH INC
  • EP4295328B1 patent drawingFigure 1
  • EP4295328B1 patent drawingFigure 2
  • EP4295328B1 patent drawingFigure 3

AI summary

Characteristics of a user's movement are evaluated based on performance of activities by a user within a field of view of a camera. Video data representing performance of a series of movements by the user is acquired by the camera. Pose data is determined based on the video data, the pose data representing positions of the user's body while performing the movements. The pose data is compared to a set of existing videos that correspond to known errors to identify errors performed by the user. The errors may be used to generate scores for various characteristics of the user's movement. Based on the errors, exercises or other activities to improve the movement of the user may be determined and included in an output presented to the user.