透過您的圖書館登入
IP:52.90.40.84
  • 學位論文

KinChat: 在聊天室中使用體感裝置及視覺化技術增 加基於特徵點的臉部表情

KinChat: Augmenting Text Chat with De-Identified Feature-based Visualization of Facial Expression

指導教授 : 王浩全

摘要


在人們透過電腦媒介傳播的方式。例如文字聊天訊息來相互溝通時。有時會 因為無法以非言語的方式來傳達意旨。例如聲調、表情等。造成雙方對於對話有 所誤解。然而。視訊溝通雖然提供了如臉部表情、手勢、外貌等更豐富的訊息。 但同時地。這些額外資訊有可能造成隱私方面的疑慮。或者對於個人外觀的偏見。 另一方面。在團體聊天室中。每個人的視訊畫面同時出現在單一視窗中。這使得 團體聊天室中的成員必須付出額外的心思與注意力。才能觀察每個人的表情。 在本論文中。我們提出了一個將文字聊天與基於特徵點所呈現的臉部表情結 合的方案。我們將這個系統命名為 KinChat。KinChat 利用 Kinect 的體感偵測 功能來追蹤並呈現使用者的臉部表情。在本論文中。我們做了兩個實驗。第一個 實驗在驗證基於特徵點的臉部表情的隱私性。以及臉部表情傳達的準確性。在第 二個實驗中。我們將 KinChat 與傳統的視訊、純文字聊天來比較。了解其對使 用者的影響。實驗結果顯示。KinChat 能在提供更佳的隱私保護的情況下。保有 一定程度的臉部表情辨識的準確性。此外。在實際對話中。實驗顯示 KinChat 能 降低使用者的社交焦慮。同時提升他們對於整個對話內容的理解程度。除了這兩 個實驗外。我們也嘗試將多人的表情合成為單一表情。希望藉此傳達團體整體的 情緒。在本論文的最後。我們亦進行了一個初步實驗以測試此功能在團體溝通中的效用。

並列摘要


In computer-mediated communication (CMC), the lack of non-verbal visual cues in text communication can cause misunderstanding, as there may not be enough social cues (e.g., facial expression) for one to determine how their partners feel and whether they understand the messages. However, video-mediated communication with rich social cues including how one looks like may raise privacy concerns and social biases. Participants of group conversation also face the difficulty in managing rich visual cues produced by multiple group members. In this thesis, we propose an interaction technique namely KinChat that combines text chat with feature-based visualization of facial expression that uses motion sensor and 2D graphics to trace and represent interlocutors’ faces during conversations. We conducted two exploratory studies to examine the utilities of this feature-based visualization on de-identification and awareness of facial expression, as well as the broader impact on the outcomes of interactive communication. The results indicate that feature-based visualization of facial expression can preserve both awareness of facial expression and privacy, and can lead to improved understanding and reduced communication anxiety when applied to communication settings. In addition, we designed an extra application, which represents general emotion and social atmosphere of a group with a single synthesized visualization of multiple users’ faces. We also conducted a pilot study to examine this application design.

參考文獻


1. Argyle, M. Non-verbal communication in human social interaction. Non-Verbal Communication 12, 443.
2. Bailenson,J.N., Yee,N., Merget,D. and Schroeder,R. The Effect of Behavioral Realism and Form Realism of Real-Time Avatar Faces on Verbal Disclosure, Nonverbal Disclosure, Emotion Recognition, and Copresence in Dyadic Interaction. Presence: Teleoperators and Virtual Environments 15, 4 (2006) 359–372
3. Bassili, J. N. Emotion Recognition: The Role of Facial Movement and the Relative Importance of Upper and Lower Areas of the Face. Journal of personality and social psychology 37, 11 (1979), 2049-2058
4. Baylor,A.L. Promoting motivation with virtual agents and avatars: role of visual presence and appearance. Philosophical Transactions of The Royal Society B 364, 1525 (2009), 3559-3565.
7. Booth-Butterfield,S.,&Gould,M.The Communication Anxiety Inventory: Validation of State-and-Context Communication Apprehension. Communication Quarterly 34 (1986), 194-205.

延伸閱讀