The dataset includes 1000 labelled Google Street View images for building façade elements detection.
Doors and windows are labelled in each image based on the YOLO v5 PyTorch format.
The dataset are splitted as train, validatioan, and test.
The dataset can be downloaded from following link:
https://drive.google.com/file/d/1CBHbUsdOAIOCEYi4oAB03senfi_9-WpP/view?usp=drive_link
If you use the dataset, please cite these publications:
Sezen, G., Cakir, M., Atik, M. E., & Duran, Z. (2022). Deep learning-based door and window detection from building façade. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 43, 315-320.
Isiler, M., Yanalak, M., Atik, M. E., Atik, S. O., & Duran, Z. (2023). A Semi-Automated Two-Step Building Stock Monitoring Methodology for Supporting Immediate Solutions in Urban Issues. Sustainability, 15(11), 8979. MDPI AG. Retrieved from http://dx.doi.org/10.3390/su15118979