SFS-A68: a dataset for the segmentation of space functions in apartment buildings
This work addresses the problem of automated space function classification for building analysis, but it is incremental as it primarily introduces a new dataset rather than a novel method.
The authors tackled the lack of deep learning image segmentation methods for space function classification in building models by creating the SFS-A68 dataset from 68 apartment building layouts, and they demonstrated its applicability through experimental training and testing.
Analyzing building models for usable area, building safety, or energy analysis requires function classification data of spaces and related objects. Automated space function classification is desirable to reduce input model preparation effort and errors. Existing space function classifiers use space feature vectors or space connectivity graphs as input. The application of deep learning (DL) image segmentation methods to space function classification has not been studied. As an initial step towards addressing this gap, we present a dataset, SFS-A68, that consists of input and ground truth images generated from 68 digital 3D models of space layouts of apartment buildings. The dataset is suitable for developing DL models for space function segmentation. We use the dataset to train and evaluate an experimental space function segmentation network based on transfer learning and training from scratch. Test results confirm the applicability of DL image segmentation for space function classification. The code and the dataset of the experiments are publicly available online (https://github.com/A2Amir/SFS-A68).