Share

Recommended by the IIR

Using Deep Learning in Real-Time for Clothing Classification with Connected Thermostats

Author(s) : MEDINA A., MÉNDEZ J. I., PONCE P., PEFFER T., MEIER A., MOLINA A.

Type of article: Periodical article

Summary

Thermal comfort is associated with clothing insulation, conveying a level of satisfaction with the thermal surroundings. Besides, clothing insulation is commonly associated with indoor thermal comfort. However, clothing classification in smart homes might save energy when the end-user wears appropriate clothes to save energy and obtain thermal comfort. Furthermore, object detection and classification through Convolutional Neural Networks has increased over the last decade. There are real-time clothing garment classifiers, but these are oriented towards single garment recognition for texture, fabric, shape, or style. Consequently, this paper proposes a CNN model classification for the implementation of these classifiers on cameras. First, the Fashion MNIST was analyzed and compared with the VGG16, Inceptionvv4, TinyYOLOv3, and ResNet18 classification algorithms to determine the best clo classifier. Then, for real-time analysis, a new dataset with 12,000 images was created and analyzed with the YOLOv3 and TinyYOLO. Finally, an Azure Kinect DT was employed to analyze the clo value in real-time. Moreover, real-time analysis can be employed with any other webcam. The model recognizes at least three garments of a clothing ensemble, proving that it identifies more than a single clothing garment. Besides, the model has at least 90% accuracy in the test dataset, ensuring that it can be generalized and is not overfitting.

Available documents

Format PDF

Pages: 28 p.

Available

Free

Download

Details

  • Original title: Using Deep Learning in Real-Time for Clothing Classification with Connected Thermostats
  • Record ID : 30029420
  • Languages: English
  • Subject: Technology
  • Source: Energies - 15 - 5
  • Publishers: MDPI
  • Publication date: 2022/03
  • DOI: http://dx.doi.org/https://doi.org/10.3390/en15051811

Links


See other articles in this issue (3)
See the source

Indexing