人臉特徵更換與形態轉換系統

Facial feature replacement and morphing systems

陳忠勇(Chung-Yung Chen)

指導教授 ： 林慧珍

淡江大學/工學院/資訊工程學系碩士班/碩士(2010年)

https://doi.org/10.6846/TKU.2010.00383

摘要

主動式外觀模型(active appearance model, AAM)是近年來受到關注的一個以模型為基礎的方法，由於它能很成功地用來建立可變形樣板模型(deformable template model)，對具多變性物件的影像有良好的闡釋功能，因而廣泛地被用來解析人臉影像。AAM 除了本身有很好的應用之外，其處理過程使用的模組也非常好用。
本篇論文中提出一個人臉特徵更換系統及一個人臉影像形態轉換系統，兩系統均利用控制點與AAM在變形過程採用到的三角剖分演算法(triangulation algorithm)做人臉特徵裁貼(facial feature pasting)。
實驗結果顯示，本文所提的人臉特徵更換法與人臉影像形態轉換法均能得到頗為自然的結果，對影像中的人臉形狀、大小或旋轉角度之差異完全沒有限制，而且可根據使用者的喜好完成不同的結果，具高度的彈性。

關鍵字

主動式外觀模型 ； 三角剖分演算法 ； 人臉特徵裁貼 ； 形態轉換

並列摘要

An AAM (Active Appearance Model) is a model-based approach for the interpretation of face images has received a great deal of attention in recent years. Due to its capability of ‘interpreting’ the images containing objects with large variability in terms of a compact set of model parameters, this method has been widely adopted to ‘explaining’ the appearance of human faces. AAM and its auxiliary modules as well are very applicable in many tasks.
In this thesis, we proposed a facial feature replacement system and a facial morphing system both using the triangulation algorithm to perform facial feature replacement or blending associated with control points.
The experimental results show that both systems are flexible to use and provide quite natural composite images. In addition, the systems have no limit to the shape, size, and plane rotation of the faces.

並列關鍵字

active appearance model(AAM) ； triangulation algorithm ； facial feature pasting ； morphing

參考文獻

[1] T. F. Cootes, C. J. Taylor, D. Cooper, and G. J. Edwards, “Active Shape Model – Their Training and Application,” Computer Vision and Image Understanding, Vol. 61, No. 1, pp. 38-59, Jan. 1995.

[2] T. F. Cootes, G. J. Edwards, and C. J. Taylor, “Active Appearance Models,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 23, No. 6, pp. 681-685, June 2001.

[3] J.G. Bosch, S.C. Mitchell, B.P.F. Lelieveldt, F. Nijland, O. Kamp, M. Sonka, and J.H.C Reiber, “Automatic segmentation of echocardiographic sequences by active appearance motion models,” IEEE Transactions on Medical Imaging, Vol. 21, No. 11, pp. 1374-1383, Nov. 2002.

[4] G. Xinbo, S. Ya, L. Xuelong, and T. Dacheng, “A Review of Active Appearance Models,” IEEE Transactions on Systems, Man, and Cybernetics - Part C: Applications and Reviews, Vol. 40, No. 2, pp. 145-158, March 2010.

[5] J. Hays and A. A. Efros, “Scene Completion Using Millions of Photographs,” ACM Transactions on Graphics, Vo1. 51, Issue 10, pp. 87-94, 2007.