Cooking Motion Estimation via Hand Utensil Overlap

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional technologies are unable to effectively capture cooking motions by combining human motion recognition with cooking utensil recognition.

Innovation Solution

A cooking motion estimation device that uses posture recognition technology to identify hand regions and object recognition technology to identify cooking utensil regions, allowing for the estimation of cooking motions when the hand and utensil regions overlap.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional posture recognition technology is used alone, then human motion can be recognized, but cooking utensil usage cannot be determined

Engineering Contradiction:

Improvehand region detection accuracyVSAvoidcooking motion information

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The patent combines posture recognition technology with object recognition technology to simultaneously detect hand regions and cooking utensils. By merging these two recognition systems, the patent achieves both accurate hand detection and cooking motion determination, resolving the limitation of using posture recognition alone.

Inventive Principle:

Principle #5Merging (Combining)

2Measurement precision

If conventional object recognition technology is used alone, then cooking utensils can be identified, but hand motion and cooking action cannot be determined

Engineering Contradiction:

Improvecooking utensil detection accuracyVSAvoidcooking motion information

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The patent merges object recognition technology with posture recognition technology to simultaneously identify cooking utensils and detect hand motions. This combination enables the system to determine both utensil presence and cooking actions, overcoming the limitation of using object recognition alone.

Inventive Principle:

Principle #5Merging (Combining)

3Measurement precision

If manual observation of cooking behavior is performed, then detailed cooking motions can be recorded, but time and effort requirements increase significantly

Engineering Contradiction:

Improvecooking behavior analysis accuracyVSAvoidtime for behavior observation

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces manual observation with automated image processing technology. By using computers to automatically detect hand regions, identify cooking utensils, and determine cooking motions from video frames, the system eliminates the need for time-consuming manual observation while maintaining or improving measurement precision.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

4Productivity

If automated image processing is used, then analysis speed increases, but the ability to accurately recognize cooking motions decreases

Engineering Contradiction:

Improvecooking behavior analysis speedVSAvoidcooking motion recognition accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent introduces an intermediary approach that uses automated image processing to detect basic features (hand regions, utensil regions) and then applies logical rules to determine cooking motions. This intermediary processing layer maintains high analysis speed while achieving accurate cooking motion recognition by combining automated detection with rule-based reasoning.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20250201026A1Cooking motion estimation device, cooking motion estimation method, and cooking motion estimation program

Publication Date: 2025.06.19 AJINOMOTO CO INC

AI summary

Coordinates of a joint point are identified, and a hand region that is a coordinate region of a hand is estimated, for each of video frames that constitute a cooking behavior video, based on posture recognition technology. A cooking utensil region that is a coordinate region of a cooking utensil is identified for each of the video frames that constitute the cooking behavior video, based on object recognition technology. When the hand region and the cooking utensil region overlap, a cooking motion for each of the video frames is estimated from a type of the cooking utensil.