Representing Hierarchies of Visual Regard in Eye-Tracking Analysis
Bibliographic Data
| ID | 3446175 |
|---|---|
| Authors | Eugene Han (0000-0003-1927-3677, Eugene Han, Lehigh University, Department of Art, Architecture, and Design, 113 Research Drive, Building C, Bethlehem, PA 18015, U.S.A. Email: [email protected]., corresponding author) |
| Year | 2022 |
| Volume | 55 |
| Issue | 1 |
| Pages | 51-56 |
| Publication date | 2022-02-23 |
| Open Access | No |
| Type | ARTICLE |
| Venue | Leonardo (JOURNAL) |
| Journal identifiers | ISSN: 0024-094X • E-ISSN: 1530-9282 |
| Publisher | The MIT Press (PUBLISHER • US) |
| DOI | 10.1162/leon_a_02096 |
| OpenAlex | W3176640682 |
| Language | EN |
| Citations received | 1 |
| References cited | 8 |
The author developed a method for representing data from eye-tracking recordings: a form of graphical analysis that illustrates hierarchical densities of visual regard without obscuring the original pictorial stimulus. Across three different case studies, subjects' fixation patterns were used to propagate Voronoi generating points. Integrating both fixation locations and their respective dwell times, randomized Gaussian distribution provided a technique to augment Voronoi generating seeds and enhance graphical resolution. Color pixel values were then used to fill in resultant Voronoi cells in relation to color values provided by the original stimulus. This analysis allowed for effective differentiation of viewing behaviors between different subjects, with emphasis on a subject's attentional distribution rather than on graphic icons
Biology · Cognitive psychology · Computer graphics (images) · Computer vision · Eye tracking · Fixation (population genetics) · Gaussian · Geometry · Pattern recognition (psychology) · Physics · Pixel · Stimulus (psychology) · Voronoi diagram · Artificial Intelligence · Computer Science · Gaze Tracking and Assistive Technology · Mathematics · Psychology · Visual Attention and Saliency Detection · Visual perception and processing mechanisms
| Citations per year | 0.33 |
|---|---|
| Citation span | 2023 - 2023 (1) |
| Highly cited | No |