Window to Wall Ratio Detection using SegFormer

TL;DR

This paper presents a method to estimate Window to Wall Ratios (WWR) from street view images using semantic segmentation, enabling better building performance modeling without extensive data collection.

Contribution

It adapts existing computer vision segmentation techniques to accurately predict WWR from external images, addressing data scarcity in architectural modeling.

Findings

Semantic segmentation can effectively estimate WWR from street images.

The method improves building performance assessments by providing accessible WWR data.

Potential for integrating computer vision into architectural design workflows.

Abstract

Window to Wall Ratios (WWR) are key to assessing the energy, daylight and ventilation performance of buildings. Studies have shown that window area has a large impact on building performance and simulation. However, data to set up these environmental models and simulations is typically not available. Instead, a standard 40% WWR is typically assumed for all buildings. This paper leverages existing computer vision window detection methods to predict WWR of buildings from external street view images using semantic segmentation, demonstrating the potential for adapting established computer vision technique in architectural applications