3D Exercise Analysis System for Gymnastics Position Holding

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Judges face difficulties in accurately evaluating holding positions in gymnastic exercises due to the challenge of identifying and reviewing the duration of position holding, leading to potential score deductions and inaccuracies in the evaluation process.

Innovation Solution

A display method utilizing 3D sensing data and element dictionary data to identify and highlight the time portion corresponding to holding positions within a series of exercises, enabling precise evaluation of position holding requirements through a processor-driven system.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If judges review recorded video to evaluate holding positions, then evaluation can be performed, but it is difficult to accurately identify and review the duration of position holding

Engineering Contradiction:
Improveevaluation accuracyVSAvoidposition holding identification difficulty
Core Design Contradiction:
Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The system performs preliminary analysis of the recorded video to automatically detect and identify holding positions before the judge reviews the footage. By pre-processing the video data to detect when the athlete maintains a specific position for the required duration, the system prepares evaluation-ready information that eliminates the difficulty of manual identification during judge review

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The invention introduces an intermediary system between the recorded video and the judge's evaluation. This intermediary automatically analyzes video frames, detects holding positions, calculates duration, and presents processed information to the judge, thereby resolving the difficulty of accurate manual detection while maintaining evaluation accuracy

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If judges manually review video to determine holding position duration, then score evaluation can be performed, but time-consuming and prone to errors

Engineering Contradiction:
Improvescore determination accuracyVSAvoidevaluation time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The invention replaces the manual mechanical review process with an automated computer-based system. Instead of judges manually watching and timing holding positions, the system uses automated video analysis algorithms to detect positions, measure duration, and determine scores, thereby eliminating time loss and improving reliability

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

Solution Approach 2:

The system performs self-service evaluation by automatically analyzing the video data, identifying holding positions, calculating their duration, and generating score recommendations without requiring extensive manual intervention from judges. This automation significantly reduces evaluation time while maintaining or improving score determination accuracy

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS11484766B2Display method, information processing device, and computer-readable recording medium
Publication Date: 2022.11.01 FUJITSU LTD
  • US11484766B2 patent drawing
  • US11484766B2 patent drawing
  • US11484766B2 patent drawing

AI summary

A display method executed by a processor includes: acquiring an identification result relating to any one of a plurality of elements included in a series of exercise and a position of an athlete that performs the elements based on 3D sensing data that is acquired by sensing the series of exercise by the athlete of a sport scored by a judge and an element dictionary data in which characteristics of an element and a position of the sport scored by a judge are defined; identifying any one of an element and a position that is being displayed in a 3D model video based on the 3D sensing data corresponding to the series of exercise, based on the identification result; and identifying, for the any one of the element and the position being displayed, time information corresponding to a boundary with any one of a next element and a next position.