student profile: Mr Dheyaa Hussein


Thesis work

Thesis title: A methodological framework for the three-dimensional evaluation of visual complexity in building fa�e based on users� preferences

Supervisors: Somwrita SARKAR , Peter ARMSTRONG

Thesis abstract:

This work aims to provide a methodology to assess the perceptual impact of visual complexity of building facades on preferences of people. In existing literature, theories on the evaluation of visual complexity are only verified on limited instances, and results on the relationship between visual complexity and preferences on a broad scope of application are not evident. How and what aspects of the visual environment affect the perception of visual complexity are not entirely clear. Thus, scientifically examined propositions that link the impact of aspects of visual complexity in building facades and the perception of users may enrich and inform future professional design practice.
The research identifies four variables that incorporate the impact of visual complexity. These variables are the number of design elements, the variation in their position and colour, and the height of the building facade. It introduces the concept of vertices as an atomic indicator on which the physical measurement of the four variables is built. The study designs seven experiments; each involves the parametric generation of a group of scenes (building facades) that systematically explore different combinations of the interactions between the four variables and collects participants’ preferences towards them. The research developed an online application, which offers a real-time interactive three-dimensional visualisation of these scenes and measures the four variables of visual complexity in them. Through an online questionnaire, it then collects participants’ preferences towards them using pairwise comparisons. Finally, the application uses statistical analysis to determine the impact of visual complexity via the variables on people’s preferences.
The research offers a quantitative lens on the effect of visual complexity as a comprehensive phenomenon through the statistical demonstration of the impact of the combinations between its variables. It also shows the importance of each variable compared to the others. The research explains the relationships between visual preferences expressed by people on a wide range of the values of visual complexity and its variables. It demonstrates that these variables can be systematically measured, and there are clear relationships between them and preferences with a certain direction and magnitude. They make possible the statistical prediction of preference based on the value of variables using a fitted regression model.
While the thesis presents the methodology developed on synthetically generated scenes, in a real situation, participants could be residents of an area where a new design is being proposed. The methodology only demands that the design being presented for preference evaluation is created using 3D software so that primary measurements of the design variables are enabled. This methodology can be used by planning authorities and design firms as an objective way to evaluate the preferences of residents based on systematic variations of visual complexity of new designs.

Selected publications

Download citations: PDF RTF Endnote


  • Hussein, D., Armstrong, P. (2016). Building an Arithmetic Model to Assess Visual Consistency in Townscape. IJCEE: International Journal of Civil & Environmental Engineering, 10(4), 457-464.


  • Hussein, D., Armstrong, P. (2016). Building an Arithmetic Model to Assess Visual Consistency in Townscape. IJCEE: International Journal of Civil & Environmental Engineering, 10(4), 457-464.

Note: This profile is for a student at the University of Sydney. Views presented here are not necessarily those of the University.