@article{35839,
  keywords = {Thermal comfort, Clothing classifier, CNN models, Connected thermostat},
  author = {Adán Medina and Juana Isabel Méndez and Pedro Ponce and Therese Peffer and Alan K Meier and Arturo Molina},
  title = {Using Deep Learning in Real-Time for Clothing Classification with Connected Thermostats},
  abstract = {<p style="text-align: left;">Thermal comfort is associated with clothing insulation, conveying a level of satisfaction<br />with the thermal surroundings. Besides, clothing insulation is commonly associated with indoor<br />thermal comfort. However, clothing classification in smart homes might save energy when the end-<br />user wears appropriate clothes to save energy and obtain thermal comfort. Furthermore, object<br />detection and classification through Convolutional Neural Networks have increased over the last<br />decade. There are real-time clothing garment classifiers, but these are oriented towards single<br />garment recognition for texture, fabric, shape, or style. Consequently, this paper proposes a CNN<br />model classification for the implementation of these classifiers on cameras. First, the Fashion<br />MNIST was analyzed and compared with the VGG16, Inceptionvv4, TinyYOLOv3, and ResNet18<br />classification algorithms to determine the best clo classifier. Then, for real-time analysis, a new<br />dataset with 12,000 images was created and analyzed with the YOLOv3 and TinyYOLO. Finally,<br />an Azure Kinect DT was employed to analyze the clo value in real-time. Moreover, real-time<br />analysis can be employed with any other webcam. The model recognizes at least three garments<br />of a clothing ensemble, proving that it identifies more than a single clothing garment. Besides, the<br />model has at least 90% accuracy in the test dataset, ensuring that it can be generalized and is not<br />overfitting.</p>},
  year = {2022},
  journal = {Energies},
  volume = {15},
  pages = {1811},
  month = {03/2022},
  url = {https://www.mdpi.com/1996-1073/15/5/1811},
  doi = {10.3390/en15051811},
  language = {eng},
}