Deep Umbra: A Generative Approach for Sunlight Access Computation in Urban Spaces

TL;DR

Deep Umbra introduces a global-scale, conditional GAN framework for quantifying accumulated sunlight access and shadows in urban spaces. By translating height tiles (b) informed by latitude (l) and season (t) into accumulated shadow maps (a), the approach achieves real-time-like performance and broad transferability across cities. The authors demonstrate a ~6× speedup versus the prior state of the art, publish the Global Shadow Dataset covering over 100 cities, and validate transferability with unseen cities while delivering qualitative and case-study insights for urban planning. Limitations include the absence of greenery modeling and reliance on height data, with future work targeting tree effects and GIS integration to support resilient urban design.

Abstract

Sunlight and shadow play critical roles in how urban spaces are utilized, thrive, and grow. While access to sunlight is essential to the success of urban environments, shadows can provide shaded places to stay during the hot seasons, mitigate heat island effect, and increase pedestrian comfort levels. Properly quantifying sunlight access and shadows in large urban environments is key in tackling some of the important challenges facing cities today. In this paper, we propose Deep Umbra, a novel computational framework that enables the quantification of sunlight access and shadows at a global scale. Our framework is based on a conditional generative adversarial network that considers the physical form of cities to compute high-resolution spatial information of accumulated sunlight access for the different seasons of the year. We use data from seven different cities to train our model, and show, through an extensive set of experiments, its low overall RMSE (below 0.1) as well as its extensibility to cities that were not part of the training set. Additionally, we contribute a set of case studies and a comprehensive dataset with sunlight access information for more than 100 cities across six continents of the world. Deep Umbra is available at https://urbantk.org/shadows.

Figures (12)

• Figure 1: Left: Example of a single timestep shadow. Right: accumulated shadows. By accumulating shadows between a time range, we can comprehensively analyze the impact of buildings on the sunlight access of public spaces. In the illustration, $P_1$ is 100% of the time under shadow, $P_2$ and $P_3$ 50%, and $P_4$ 25%. The rightmost image shows the resulting accumulation when considering a time range between 10 AM and 4 PM.
• Figure 2: Overview the different components in the Deep Umbra framework. Step 1, left: The preprocessing component is responsible for extracting building height information from OpenStreetMap OpenStreetMap (a) and generate $512\times 512$ tiles with this information (b). Step 2, right: The preprocessed data is then used to train a conditional generative adversarial network to map height $\rightarrow$ accumulated shadow, also taking into account season of the year and latitude of the tile (c). Cities used in the training set are shown as *B on the map, and cities used for testing are shown as *B; gt refers to ground truth data and ge GAN generated data.
• Figure 3: Example where shadows are cast across tiles. In (b), the two small squares highlight areas where shadows from neighboring buildings are cast onto the center tile. This is further highlighted for different seasons in (a) and (c), with vertical black lines indicating the boundaries between tiles. To account for accumulated shadows from neighboring tiles, during training, a $256\times256$ height tile is padded with parts of adjacent tiles, resulting in a $512\times512$ tile (larger purple square in (b)). Model results are cropped back to the original size of $256\times256$ (smaller green square in (b)).
• Figure 4: Distribution of average tile heights for the train and test sets.
• Figure 5: Results of ablation study using U-Net and ResNet generator architecture and $L_1$ loss function. RMSE and SSIM scores with respect to ground truth are highlighted below each image (for SSIM, higher values are better). Note that ResNet architecture produces significantly better results, especially for longer shadows that are distant from the building. Top: Highlights with areas with longer shadows. Bottom: Zoomed areas.