A method for using street view imagery to auto-extract window-to-wall ratios and its relevance for urban-level daylighting and energy simulations

Szcześniak, Jakub; Ang, Yu Qian; Letellier-Duchesne, Samuel; Reinhart, Christoph

doi:10.1016/j.buildenv.2021.108108

Cited by 36 publications

(24 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It could also be combined with advanced urban energy modeling software with 3D capabilities to provide stochastic boundary conditions. What we also wish to highlight is its potential for direct handling of real-world images, thanks to the rapid development of urban scenario segmentation applications in recent years [43]. The label mask obtained from the semantic segmentation is highly similar to what we need.…”

Section: Discussionmentioning

confidence: 89%

SolarGAN: Synthetic Annual Solar Irradiance Time Series on Urban Building Facades via Deep Generative Networks

Zhang¹,

Schlueter²,

Waibel³

2022

Preprint

View full text Add to dashboard Cite

A generative model using simple fisheye images as inputs to predict solar irradiance time series on building facades in urban environments• Fisheye images as categorical shading masks are captured from opensource LOD 1 urban geometry, but in principle may also stem from real photographs• Model architecture allows for generation of stochastic ensembles of annual hourly time series, as well as good consistency to the deterministic data• Developed a network architecture combining Variational Autoencoder (VAE) and Generative Adversarial Networks (GAN)• Developed model addresses inherent uncertainty of intermittent solar energy and allows parametrically altering urban environmental features

show abstract

Section: Discussionmentioning

confidence: 89%

SolarGAN: Synthetic Annual Solar Irradiance Time Series on Urban Building Facades via Deep Generative Networks

Zhang¹,

Schlueter²,

Waibel³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The annotation rule follows uniform Haussmanian-style grammar, i.e., all windows are annotated as rectangles, even though some of them are arc-shaped. This dataset has been widely used to evaluate window detection or fac ¸ade segmentation approaches [28], [61], [62], [3].…”

Section: B Datasetmentioning

confidence: 99%

Cross Field-Based Segmentation and Learning-Based Vectorization for Rectangular Windows

Zhuo

Tian

Fraundorfer

2023

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Detection and vectorization of windows from building fac ¸ades are important for building energy modeling, civil engineering, and architecture design. However, current applications still face the challenges of low accuracy and lack of automation.In this paper we propose a new two-steps workflow for window segmentation and vectorization from fac ¸ade images. First, we propose a cross field learning-based neural network architecture, which is augmented by a grid-based self-attention module for window segmentation from rectified fac ¸ade images, resulting in pixel-wise window blobs. Second, we propose a regression neural network augmented by Squeeze-and-Excitation (SE) attention blocks for window vectorization. The network takes the segmentation results together with the original fac ¸ade image as input, and directly outputs the position of window corners, resulting in vectorized window objects with improved accuracy. In order to validate the effectiveness of our method, experiments are carried out on four public fac ¸ades image datasets, with results usually yielding a higher accuracy for the final window prediction in comparison to baseline methods on four datasets in terms of IoU score, F1 score, and pixel accuracy.

show abstract

“…Street-level window-to-wall ratios (WWRs) have long played a critical role in assessing the visual qualitythe visual stimuli that shape our perceptions, attitudes, and general views (Sheikh & Bovik, 2006)of urban street façades (Li et al, 2020). The measurement of WWRs at an urban scale has been conducted only for building or urban energy assessments (Shi et al, 2020;Szcześniak et al, 2022).…”

Section: Eyes On the Street: Assessing Window-to-wall Ratios In Googl...mentioning

confidence: 99%

“…By stitching the images together, GSV images can create continuous building façade experiences of a streetscape. Thus, the GSV image library has been proposed as an effective potential data source for urban assessments (Rundle et al, 2011;Rzotkiewicz et al, 2018), such as public open space audits (Edwards et al, 2013;Taylor et al, 2011), neighborhood environmental audits (Charreire et al, 2014;Rundle et al, 2011;Odgers et al, 2012;Griew et al, 2013), and building energy audits (Shi et al, 2020;Szcześniak et al, 2022).…”

Section: Related Workmentioning

confidence: 99%

SIGraDi 2022. Critical Appropriations

Z.¹

2023

View full text Add to dashboard Cite

SIGraDi 2022. Critical Appropriations es la XXVI Conferencia Internacional de la Sociedad Iberoamericana de Gráfica Digital, organizada por la Facultad de Arquitectura de la Universidad Peruana de Ciencias Aplicadas (UPC) y SIGraDi. Este volumen contiene los artículos seleccionados por el comité científico de SIGraDi, los cuales fueron presentados en línea entre el 9 y 11 de noviembre de 2022, además, a partir de estos trabajos se desarrollaron10 talleres y un taller para doctorandos. SIGraDi 2022. Apropiaciones Críticas se organizó en 7 tópicos y 30 subtópicos, como culturas de la programación, robótica y fabricación digital, BIM, diseño bio inspirado, educación digital. Una publicación de difusión e intercambio de ideas sobre Arquitectura, Diseño, Arte y disciplinas asociadas.

show abstract

A method for using street view imagery to auto-extract window-to-wall ratios and its relevance for urban-level daylighting and energy simulations

Cited by 36 publications

References 14 publications

SolarGAN: Synthetic Annual Solar Irradiance Time Series on Urban Building Facades via Deep Generative Networks

SolarGAN: Synthetic Annual Solar Irradiance Time Series on Urban Building Facades via Deep Generative Networks

Cross Field-Based Segmentation and Learning-Based Vectorization for Rectangular Windows

SIGraDi 2022. Critical Appropriations

Contact Info

Product

Resources

About