PhD Dissertation, EPFL
Daylit architecture is perceived as a dynamic luminous composition, yet existing performance metrics most often evaluate natural illumination for its ability to adequately illuminate a two-dimensional task surface while avoiding glare-based discomfort. This limited task-driven approach places a disproportionate emphasis on surface illumination and glare-based discomfort and ignores the likelihood that contrast can provide a positive impact on our visual perception. Existing studies have attempted to link subjective ratings of composition and preference to simple global contrast metrics without reaching a strong consensus. More robust metrics have been developed in computational graphics and vision research, but have not been applied to studies in qualitative lighting research. As daylight-driven visual effects in architecture are influenced by dynamic sky conditions, this research introduces an experimental method through which ratings of daylight composition in architectural renderings are then compared to existing contrast metrics across a series of annual moments. This research seeks to identify which quantitative metrics correlate to subject ratings of visual interest, and introduces a quantitative measure which can be used as a novel prediction model for characteristics of visual interest in dynamic renderings of daylit architecture.