詳細検索結果
以下の条件での結果を表示する: 検索条件を変更
クエリ検索: "日本似顔絵師協会"
2件中 1-2の結果を表示しています
  • 中洲 俊信, 池 司, 山内 康晋, 斉藤 孝浩
    ヒューマンインタフェース学会論文誌
    2018年 20 巻 1 号 115-124
    発行日: 2018/02/25
    公開日: 2018/02/25
    ジャーナル フリー
    A user interface (UI) which enables users to operate digital products remotely without touching displays or keyboards is required in such a case as their hands are wet. We developed a hands-free UI which realized remote gesture operation by using a wristband-type wearable sensor. This UI recognizes the posture change of user’'s arm and "tapping" gesture, and introduces them to pointing and clicking respectively. In this paper we evaluate the usability of this UI adapted to kitchen display system. We had 10 users complete menu manipulation tasks by gesture and extracted usability problems on gesture operation, one of which was the menu selection error due to sensor shake in tapping gesture. Moreover, we propose a novel click position estimation technique to reduce the error by click position modification based on the mouse cursor's trace. The evaluation results confirmed that our method was effective in gesture control.
  • 中洲 俊信, 山地 雄土, 柴田 智行, 井本 和範
    ヒューマンインタフェース学会論文誌
    2016年 18 巻 3 号 141-152
    発行日: 2016/08/25
    公開日: 2019/07/01
    ジャーナル フリー

    We propose a novel interactive retrieval system which enables a user to intuitively retrieve materials. Our system needs a user to sketch an abstract of the target material as search query, which indicates component layout, color of components and class of components; text, figure or table. In general, the expression of hand-written query is dependent on user's sense or appearance of the target material. To realize a system which enables a user to search freely and intuitively, it is necessary for the system to recognize a wide variety of sketch expressions. To define expressions which our system should recognize, we had 165 users draw search queries in a preliminary experiment and extracted common features of sketch expressions. Based on this knowledge, we developed an interactive retrieval system that recognizes attributes of the components in hand-written query and retrieves similar materials in the aspect of recognized attributes; layout, color and class. We had 48 novice users search the target materials using our system. Results of this experiment show that our system is effective to search when a user doesn't remember detail contents, e.g. keywords, in the target material.

feedback
Top