Evaluation of Input/Output Interface using Wrinkles on Clothes (pp036-054)
         Kentaro Ueda, Tsutomu Terada, and Masahiko Tsukamoto
         doi: https://doi.org/10.26421/JDI1.1-3
Abstracts: Wearable computing has created textile-based interfaces utilizing the interaction between the user and cloth for operation, as well as the touch and the pinch input operation. The user wears and uses the device in various postures, environments, and operating positions that affect the operation speed and accuracy. However, no study has assessed such factors of touching and pinching using the same input interface. One of the textile interfaces has an input interface using wrinkles on clothes. A ridge of cloth produces a wrinkle that forms naturally on clothes, and the shape of these wrinkles can be recognized by their tactile sensations. Additionally, the act touching or pinching wrinkles does not look strange to an onlooker, which reach that wrinkles have the potential suitable for the wearable computing operation. To reveal the potential, this paper evaluates the input performance using wrinkles on clothes. We designed three touch input methods and one pinch input method for the operation using wrinkles. We implemented the input and the output device which use wrinkles and carried out four evaluations. The results indicated that the pinch input reached the highest accuracy of 98\% of four input methods after learning. The narrowing-down selection reached the fastest input time of 1.64 seconds of four methods after learning. The long press touch and the pinch input achieved the accuracy of 90\% or more in all combination of operating environments and device positions. According to the result of the wrinkle recognition, users have a high accuracy of the identification of wrinkles of 89.4\% and recognize their shape in approximately 12 seconds.
Key words:  Textile interface,Wrinkles,Wearable computing, Performance evaluation