| dc.contributor | 政治大學資訊科學系 | en_US |
| dc.contributor | 行政院國家科學委員會 | en_US |
| dc.creator (作者) | 李蔡彥 | zh_TW |
| dc.date (日期) | 2006 | en_US |
| dc.date.accessioned | 12-Nov-2012 11:01:46 (UTC+8) | - |
| dc.date.available | 12-Nov-2012 11:01:46 (UTC+8) | - |
| dc.date.issued (上傳時間) | 12-Nov-2012 11:01:46 (UTC+8) | - |
| dc.identifier.uri (URI) | http://nccur.lib.nccu.edu.tw/handle/140.119/55377 | - |
| dc.description.abstract (摘要) | 雖然電腦繪圖硬體的處理速度已有長足的進步，但由於虛擬演員的高自由度及操控介面的即時要求，在3D虛擬環境中實現動畫角色的即時控制仍是極具挑戰的課題。在本計畫中，我們以兩年的時間研究以運動計畫演算法，設計即時控制3D虛擬角色的智慧型人機介面，讓使用者能以即時互動的方式有效操控具有高自由度的動畫角色。在第一年的研究裡，我們以第一人稱視點運動控制方式為範疇，成功開發出能適應個人操控特性的線上輔助介面，並提升由電腦所產生輔助路徑的品質。在第二年的研究裡，我們嘗試將此控制方式提昇為更具挑戰的第三人稱控制法。我們根據視點是否依附在動畫角色上，設計了兩類第三人稱的控制輔助方法，即時產生能與環境互動的角色動畫。第一類是針對架設於動畫角色後方的攝影機，以分解計畫法則，產生能適應環境的人體上半身動畫。第二類是針對與動畫角色運動獨立的攝影機，以運動擷取資料庫為輔助，即時合成動畫角色多樣化的全身運動。目前這兩類技術均已能整合在一般個人電腦的操控介面上，產生能與環境互動的角色動畫。此計畫目前的研究成果已達到計畫原定目標，我們並已逐步將成果整理成論文，於各國際研討會或期刊中發表。 | - |
| dc.description.abstract (摘要) | Despite the advances in graphics hardware development, controlling an interactive 3D character is still a great challenge due to the high degrees of freedom involved in controlling a virtual character and the real-time requirement of interactive interface. In this project, we have used two years to investigate how to make use of motion planning algorithms to design an effective intelligent user interface for controlling a digital actor in real time in walkthrough applications. In the first year, we have succeeded in developing an intelligent navigation interface with a first-person view that is adaptive to user navigation behaviors. In the second year, we attempt to extend the goal to cover the more challenging third-person control mode, where the viewpoint is detached from the eyes of the virtual character. According to the fact whether the viewpoint moves with the character or not, we have developed two methods to generate motions in real time. By assuming that the camera is placed behind the character, in the first method we use a decoupled motion planning algorithm to generate compliant motions for the upper body of the character. In the second method, by making use of a library of captured motions, we attempt to synthesize versatile motions in real time for the full body of the character. Both methods have successfully been integrated into 3D user navigation interfaces operated with a mouse or keyboard on a regular desktop computer. We have achieved the goals specified in the proposal, and we are in the process of publishing the results in international conferences and journals. | - |
| dc.language.iso | en_US | - |
| dc.relation (關聯) | 應用研究 | en_US |
| dc.relation (關聯) | 學術補助 | en_US |
| dc.relation (關聯) | 研究期間:9508~ 9607 | en_US |
| dc.relation (關聯) | 研究經費:534仟元 | en_US |
| dc.subject (關鍵詞) | 智慧型3D瀏覽介面;即時運動計畫;演算法 | en_US |
| dc.title (題名) | 智慧型3D瀏覽介面中即時運動計畫演算法的設計(II) | zh_TW |
| dc.title.alternative (其他題名) | Designing Real-Time Motion Planning Algorithms for Intelligent 3d Navigation Interface(II) | en_US |
| dc.type (資料類型) | report | en |
