摘要
在靜態影像中抽取人體骨架（HB-framework）至今仍是一個重要且尚待解決的問題。在此篇論文中，我們針對正面、站姿、穿著長褲、長袖的單一人體影像，提出了一個有系統的方法以解決抽取骨架的問題。為了有效的抽取人體骨架，我們參考人體解剖學相關資料建構在此論文中所提及的人體骨架。我們所提出的抽取骨架方法，由以下三個主要部份所構成︰人臉偵測、切割衣服與褲子區域，以及嵌入骨架。人臉偵測是利用膚色及人臉模板透過樣型比對方法來達成；切割衣服及褲子區域是藉著色彩量化、紋路特徵的抽取、以及紋路相似度的計算求得；最後則是依照資訊理論中熵（entropy）的概念，透過衣服、褲子相似度的評估導出同質性（homogeneous）與異質性（inhomogeneous）資訊，藉此同質性、異質性二種資訊的輔助，便可將軀幹、手部、臀部、腳部的骨架依序套入影像中。實驗結果證實我們所提出方法的可行性。我們也透過實驗結果分析此一方法可能遭遇的問題、並討論此方法的限制性以及未來的相關應用。
並列摘要
Extracting human body framework in images is up to now an open and significant problem in general. We propose a systematic approach in which strictly follows the fundamental assumption: only one front-viewed standing human body dressing clothes with long sleeves and long trousers can be extracted from an image in this thesis. To effectively extract the human body framework, the main parts and their relationships considered in a human body are defined first from the viewpoint of anatomy. Three main steps devised in our approach to extract the human body framework are face detection, segmentation of clothes and trousers regions, and positioning all human body parts for obtaining the final human body framework. The face detection is performed by means of using face skin color and face template. Based on the determined hue quantization and color texture features, a texture similarity measure is designed for the segmentation of clothes and trousers. In accordance with the entropy concept of information theory, the homogeneous information and inhomogeneous information from similarity measurements of a human body image are introduced. Then the useful images and transformed respectively from the found homogeneous information and inhomogeneous information are combined with the defined relationships of a human body framework to position the trunk region, arm parts, the hip region and leg parts, in order. Experiments confirm the feasibility of the proposed approach, some phenomena observed from the results are discussed, the inherent limits of this approach are given, and some interesting and valuable problems are pointed out.
參考文獻
[1] V. I. Pavlovic, R. Sharma, and T. S. Huang, “Visual interpretation of hand gestures for human-computer interaction: a review, ” IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 19, No. 7, pp. 677-695, 1997.
[2] A. D. Wilson and A. F. Bobick, “Parametric hidden markov models for gesture recognition,” IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 21, No. 9, pp. 884-900, 1999.
[3] P. N. Belhumeur, J. P. Hespanha, and D. J. Kriegman, “Eigenfaces vs. fisherfaces: recognition using class specific linear projection,” IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 19, No. 7, pp. 711-720, 1997.
[4] Y. Adini, E. Moses, and S. Ullman, “Face recognition: the problem of compensating for changes in illumination direction,” IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 19, No. 7, pp. 721-732, 1997.
[5] A. Lanitis, C. J. Taylor, and T. F. Cootes, “Automatic interpretation and coding of face images using flexible models,” IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 19, No. 7, pp. 743-756, 1997.