Classification method of tactile feeling using stacked autoencoder based on haptic primary colors

Fumihiro Kato, Charith Lasantha Fernando, Yasuyuki Inoue, Susumu Tachi

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)


We have developed a classification method of tactile feeling using a stacked autoencoder-based neural network on haptic primary colors. The haptic primary colors principle is a concept of decomposing the human sensation of tactile feeling into force, vibration, and temperature. Images were obtained from variation in the frequency of the time series of the tactile feeling obtained when tracing a surface of an object, features were extracted by employing a stacked autoencoder using a neural network with two hidden layers, and supervised learning was conducted. We confirmed that the tactile feeling for three different surface materials can be classified with an accuracy of 82.0 [%].

Original languageEnglish
Title of host publication2017 IEEE Virtual Reality, VR 2017 - Proceedings
PublisherIEEE Computer Society
Number of pages2
ISBN (Electronic)9781509066476
Publication statusPublished - 2017 Apr 4
Externally publishedYes
Event19th IEEE Virtual Reality, VR 2017 - Los Angeles, United States
Duration: 2017 Mar 182017 Mar 22

Publication series

NameProceedings - IEEE Virtual Reality


Conference19th IEEE Virtual Reality, VR 2017
Country/TerritoryUnited States
CityLos Angeles


  • H.1.2
  • H.5.1 [Human-centered computing]: Virtual reality
  • H.5.2 [Information Interfaces and Presentation (e.g. HCI)]: User Interfaces (D.2.2
  • I.3.6) - Haptic I/O
  • I.5.1 [Pattern Recognition]: Models - Neural nets

ASJC Scopus subject areas

  • General Engineering


Dive into the research topics of 'Classification method of tactile feeling using stacked autoencoder based on haptic primary colors'. Together they form a unique fingerprint.

Cite this