Difference between revisions of "Three-dimensional Visual Attention Heatmap in Space: Measuring and Modelling Navigational Behaviours in Built Environment"
Abel Maciel (talk | contribs) (Created page with "Category:DCIO Category:DCIO2020 Category:DCIO Proceedings Category:Conferences DC I/O 2020 proceeding by ZEHAO QIN, AVA FATAH GEN. SCHIECK, STAM...") |
Abel Maciel (talk | contribs) (→Reference) |
||
(2 intermediate revisions by the same user not shown) | |||
Line 3: | Line 3: | ||
[[Category:DCIO Proceedings]] | [[Category:DCIO Proceedings]] | ||
[[Category:Conferences]] | [[Category:Conferences]] | ||
+ | [http://wiki.designcomputation.org/home/index.php/DC_I/O_2020#Posters DC I/O 2020 poster] by [[ZEHAO QIN]], [[AVA FATAH GEN. SCHIECK]], [[STAMATIOS PSARRAS]]. | ||
− | |||
+ | =Abstract= | ||
+ | This paper aims to provide a tool for measuring and modelling visual attention in the three-dimensional environment, which could extend the capability of spatial analysis from two-dimensionality to three-dimensionality and present a brief cognitive understanding of pedestrian viewing behaviours by analysing recorded pedestrian navigational behaviours. | ||
− | + | The overall hypothesis of this paper is that human’s navigational behaviours correlate with the visibility of objects in the space; for instance, the visibility of salient object influence pedestrian’s navigational behaviours exogenously or endogenously. Two hypotheses are stated: viewing behaviours could be measured and modelled in three-dimensionality by capturing data from reality. Visual attention heatmap was able to efficiently analyse the visual attention in three-dimensionality, and provide a satisfying result that proves that human’s viewing behaviours highly correlate with the visibility of objects in the environment. | |
+ | |||
+ | The primary methodology of identifying which elements of a visual scene are the most influential in terms of how pedestrian navigate themselves in space is visual attention. Visual attention is the subjective perceptual quality some stimuli possess within a visual scene that affects pedestrians’ potential navigational behaviour. Visual attention can be seen as a numerical representation of objects in the space that attracts pedestrian’s attention. | ||
+ | |||
+ | We measured and visualised the navigational behaviour of 493 pedestrians moving freely in an interior space, including assessments of 3,158 head orientations taken from 3,960 sequential photos taken at the location. We then analysed the visual attention heatmap generated from grouped and individual viewing and navigating behaviours and compared that data with data observed at the site. The result of visual attention heatmap was able to achieve a convincible result of visual attention in three-dimensionality within an interior space by capturing data from reality, and a connection between present spatial analysis and visual attention was established by incorporating the visual attention heatmap purposed by this research. | ||
=Keywords= | =Keywords= | ||
Line 20: | Line 26: | ||
=Reference= | =Reference= | ||
DOI: https://doi.org/10.47330/DCIO.2020.BNRH6093 | DOI: https://doi.org/10.47330/DCIO.2020.BNRH6093 | ||
+ | |||
+ | Video Presentation: https://youtu.be/_mxPiqDcmgs | ||
Full text: [https://www.designcomputation.org/dcio2020 Maciel, A. (Ed.), 2020. Design Computation Input/Output 2020, 1st ed. Design Computation, London, UK. ISBN: 978-1-83812-940-8, DOI:10.47330/DCIO.2020.QPRF9890] | Full text: [https://www.designcomputation.org/dcio2020 Maciel, A. (Ed.), 2020. Design Computation Input/Output 2020, 1st ed. Design Computation, London, UK. ISBN: 978-1-83812-940-8, DOI:10.47330/DCIO.2020.QPRF9890] |
Latest revision as of 23:39, 14 December 2020
DC I/O 2020 poster by ZEHAO QIN, AVA FATAH GEN. SCHIECK, STAMATIOS PSARRAS.
Contents
Abstract
This paper aims to provide a tool for measuring and modelling visual attention in the three-dimensional environment, which could extend the capability of spatial analysis from two-dimensionality to three-dimensionality and present a brief cognitive understanding of pedestrian viewing behaviours by analysing recorded pedestrian navigational behaviours.
The overall hypothesis of this paper is that human’s navigational behaviours correlate with the visibility of objects in the space; for instance, the visibility of salient object influence pedestrian’s navigational behaviours exogenously or endogenously. Two hypotheses are stated: viewing behaviours could be measured and modelled in three-dimensionality by capturing data from reality. Visual attention heatmap was able to efficiently analyse the visual attention in three-dimensionality, and provide a satisfying result that proves that human’s viewing behaviours highly correlate with the visibility of objects in the environment.
The primary methodology of identifying which elements of a visual scene are the most influential in terms of how pedestrian navigate themselves in space is visual attention. Visual attention is the subjective perceptual quality some stimuli possess within a visual scene that affects pedestrians’ potential navigational behaviour. Visual attention can be seen as a numerical representation of objects in the space that attracts pedestrian’s attention.
We measured and visualised the navigational behaviour of 493 pedestrians moving freely in an interior space, including assessments of 3,158 head orientations taken from 3,960 sequential photos taken at the location. We then analysed the visual attention heatmap generated from grouped and individual viewing and navigating behaviours and compared that data with data observed at the site. The result of visual attention heatmap was able to achieve a convincible result of visual attention in three-dimensionality within an interior space by capturing data from reality, and a connection between present spatial analysis and visual attention was established by incorporating the visual attention heatmap purposed by this research.
Keywords
Spatial perception, Visual attention, Space syntax, Pedestrian navigation, Viewing behaviours, Visibility graph analysis (VGA), Line of sight (LoS), Isovist, Wayfinding.
Keyphrases
visual attention (180), visual attention heatmap (126), pedestrian navigational behaviour (63), navigational behaviour (50), head orientation (40)
Topics
Architecture, Computer-Aided Design (CAD), Data Visualization and Analysis for design, Natural navigation, Simulation, Visual and Spatial Modelling.
Reference
DOI: https://doi.org/10.47330/DCIO.2020.BNRH6093
Video Presentation: https://youtu.be/_mxPiqDcmgs