Intelligent evaluation method for design education and comparison research between visualizing heat-maps of class activation and eye-movement
The evaluation of design results plays a crucial role in the development of design. This study presents a design work evaluation system for design education that assists design instructors in conducting objective evaluations. An automatic design evaluation model based on convolutional neural netw...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-10-01
|
| Series: | Journal of Eye Movement Research |
| Subjects: | |
| Online Access: | https://bop.unibe.ch/JEMR/article/view/10964 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849766470588825600 |
|---|---|
| author | Jia Jiayi Zhao Tianjiao Yang Junyu Wang Qian |
| author_facet | Jia Jiayi Zhao Tianjiao Yang Junyu Wang Qian |
| author_sort | Jia Jiayi |
| collection | DOAJ |
| description |
The evaluation of design results plays a crucial role in the development of design. This study presents a design work evaluation system for design education that assists design instructors in conducting objective evaluations.
An automatic design evaluation model based on convolutional neural networks has been established, which enables intelligent evaluation of student design works. During the evaluation process, the CAM is obtained. Simultaneously, an eye-tracking experiment was designed to collect gaze data and generate eye-tracking heat maps. By comparing the heat maps with CAM, an attempt was made to explore the correlation between the focus of the evaluation’s attention on human design evaluation and the CNN intelligent evaluation. The experimental results indicate that there is some certain correlation between humans and CNN in terms of the key points they focus on when conducting an evaluation. However, there are significant differences in background observation.
The research results demonstrate that the intelligent evaluation model of CNN can automatically evaluate product design works and effectively classify and predict design product images. The comparison shows a correlation between artificial intelligence and the subjective evaluation of human eyes in evaluation strategy. Introducing artificial intelligence into the field of design evaluation for education has a strong potential to promote the development of design education.
|
| format | Article |
| id | doaj-art-aecffa0eb58a4b1782cc3f814fe4560c |
| institution | DOAJ |
| issn | 1995-8692 |
| language | English |
| publishDate | 2024-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Eye Movement Research |
| spelling | doaj-art-aecffa0eb58a4b1782cc3f814fe4560c2025-08-20T03:04:34ZengMDPI AGJournal of Eye Movement Research1995-86922024-10-0117210.16910/jemr.17.2.1Intelligent evaluation method for design education and comparison research between visualizing heat-maps of class activation and eye-movementJia Jiayi0Zhao Tianjiao1Yang Junyu2Wang Qian3Tianjin UniversityTianjin UniversityTianjin UniversityTianjin University The evaluation of design results plays a crucial role in the development of design. This study presents a design work evaluation system for design education that assists design instructors in conducting objective evaluations. An automatic design evaluation model based on convolutional neural networks has been established, which enables intelligent evaluation of student design works. During the evaluation process, the CAM is obtained. Simultaneously, an eye-tracking experiment was designed to collect gaze data and generate eye-tracking heat maps. By comparing the heat maps with CAM, an attempt was made to explore the correlation between the focus of the evaluation’s attention on human design evaluation and the CNN intelligent evaluation. The experimental results indicate that there is some certain correlation between humans and CNN in terms of the key points they focus on when conducting an evaluation. However, there are significant differences in background observation. The research results demonstrate that the intelligent evaluation model of CNN can automatically evaluate product design works and effectively classify and predict design product images. The comparison shows a correlation between artificial intelligence and the subjective evaluation of human eyes in evaluation strategy. Introducing artificial intelligence into the field of design evaluation for education has a strong potential to promote the development of design education. https://bop.unibe.ch/JEMR/article/view/10964Eye movementEye trackingGazeAttentionDesign educationDesign evaluation |
| spellingShingle | Jia Jiayi Zhao Tianjiao Yang Junyu Wang Qian Intelligent evaluation method for design education and comparison research between visualizing heat-maps of class activation and eye-movement Journal of Eye Movement Research Eye movement Eye tracking Gaze Attention Design education Design evaluation |
| title | Intelligent evaluation method for design education and comparison research between visualizing heat-maps of class activation and eye-movement |
| title_full | Intelligent evaluation method for design education and comparison research between visualizing heat-maps of class activation and eye-movement |
| title_fullStr | Intelligent evaluation method for design education and comparison research between visualizing heat-maps of class activation and eye-movement |
| title_full_unstemmed | Intelligent evaluation method for design education and comparison research between visualizing heat-maps of class activation and eye-movement |
| title_short | Intelligent evaluation method for design education and comparison research between visualizing heat-maps of class activation and eye-movement |
| title_sort | intelligent evaluation method for design education and comparison research between visualizing heat maps of class activation and eye movement |
| topic | Eye movement Eye tracking Gaze Attention Design education Design evaluation |
| url | https://bop.unibe.ch/JEMR/article/view/10964 |
| work_keys_str_mv | AT jiajiayi intelligentevaluationmethodfordesigneducationandcomparisonresearchbetweenvisualizingheatmapsofclassactivationandeyemovement AT zhaotianjiao intelligentevaluationmethodfordesigneducationandcomparisonresearchbetweenvisualizingheatmapsofclassactivationandeyemovement AT yangjunyu intelligentevaluationmethodfordesigneducationandcomparisonresearchbetweenvisualizingheatmapsofclassactivationandeyemovement AT wangqian intelligentevaluationmethodfordesigneducationandcomparisonresearchbetweenvisualizingheatmapsofclassactivationandeyemovement |