One of the project’s central challenges was transforming qualitative spatial experiences into measurable urban indicators while maintaining comparability across very different urban contexts. The methodology combined street network extraction, image segmentation, orientation analysis, and object detection workflows in order to evaluate relationships between vegetation coverage, street orientation, pedestrian density, and environmental openness. Semantic segmentation models were used to identify trees, grass, sky visibility, and built surfaces across thousands of street images, allowing the research to quantify environmental exposure and ecological presence throughout the urban fabric. These datasets were then correlated with pedestrian activity and mobility patterns through density mapping and object detection analysis. The comparison between New York, Amsterdam, and Singapore revealed distinct spatial behaviors linked to density, orientation, vegetation distribution, and climatic adaptation. Amsterdam’s lower-density and canal-centered structure produced more continuous environmental openness, while New York’s vertical concentration and Singapore’s dense vegetation generated different forms of visual enclosure and spatial compression. By combining computational analysis with urban design thinking, the project develops a scalable framework for evaluating how morphology and environmental conditions shape everyday urban experience.