Image dataset: Year-long hourly façade photos of a university building

Marc Roca-Musach ¹, Isabel Crespo Cabillo ¹, Helena Coch ¹

⁰Architecture, Energy and Environment Research Group (AiEM) - Universitat Politècnica de Catalunya, Spain.

Data in Brief +

|

September 5, 2024

View abstract on PubMed

Summary

This study captured one year of building façade data, offering insights into user interaction with manually controlled adaptive systems. The dataset aids in developing energy simulation models and computer vision applications.

Area Of Science

Building Science
Sustainable Architecture
Human-Building Interaction

Background

Manually controlled adaptive façades are crucial for building energy efficiency and occupant comfort.
There is a significant lack of observational data on how users interact with these dynamic façade systems.
Understanding user behavior is key to optimizing building performance and energy simulations.

Purpose Of The Study

To present a comprehensive dataset of building façade interactions over one year.
To provide data for developing and validating user behavior models for adaptive façades.
To facilitate research in computer vision and machine learning for façade element analysis.

Main Methods

Acquisition of 9596 high-resolution images of building façades over a full year.
Utilizing two identical GoPro Hero 10 cameras for hourly image capture.
Systematic documentation of East and West building façades, detailing window configurations and solar protections.

Main Results

A rich dataset of 9596 façade images, capturing hourly changes.
Extracted 1.7 million individual window images for detailed analysis.
Established a foundation for analyzing user-driven façade adjustments and their impact.

Conclusions

The dataset offers unprecedented insights into real-world user-building interactions with adaptive façades.
This data is vital for enhancing the accuracy of building energy simulations and predictive models.
The generated image resources support advancements in machine learning applications for building performance monitoring.

Keywords:

Adaptive façades Computer vision Manual control Roller shutters User-façade interaction