髮型預覽系統能在理髮之前就看到理髮後的樣子，能使人更容易找到適合自己的髮型，但是髮型預覽系統需要大量的髮型資料庫。我們的自動頭髮去背可以輸入一張人臉圖片，自動對圖片作去背後只留下頭髮，故使用我們的系統即可迅速的產生髮型資料庫。
我們提出一個自動頭髮去背演算法的架構，其演算法流程是繪製頭髮線、臉部線、邊界線，再作頭部去背和臉部去背，並利用兩張去背的alpha建立alpha空間，找到品質較好的trimap，再使用trimap對原圖作去背得到結果。我們另外提出一個良好的評價函數:alpha相似度，用來評量實驗結果，也在演算法中使用。我們的ground truth是以手動的方式對原圖畫trimap產生品質最好的去背alpha，實驗結果有86.2%的相似度。

Automatic hair extraction from a given 2D image has been a challenging problem for a long time, especially when complex backgrounds and a wide variety of hairstyles are involved. This paper has made its contribution in the following two aspects. First, it proposes a novel framework that successfully combines the techniques of face recognition, outlier-aware initial stroke placement and matting to extract the desired hairstyle from an input image. Second, it defines a new comparison metric that is well suited for the alpha matte comparison. Our results show that, compared with the manually drawn trimaps for hair extraction, the proposed automatic algorithm can achieve about 86.2% extraction accuracy.

[1] Paul Viola, Michael Jones. Rapid Object Detection using a Boosted Cascade of Simple Features. Computer Vision and Pattern Recognition, 2001.
[2] Timothy F. Cootes, Christopher J. Taylor, D. H. Cooper, J. Graham. Active Shape Models-Their Training and Application. In Computer Vision and Image Understanding, volume 61(1), pages 38-59, 1995.
[3] Timothy F. Cootes, Gareth J. Edwards, Christopher J. Taylor. Active Appearance Models. In IEEE Transactions on Pattern Analysis and Machine Intelligence, volume 23(6), 2001.
[4] Irad Ben-Gal. Outlier Detection. In Data Mining and Knowledge Discovery Handbook, 2005.
[5] Edwin M. Knorr, Raymond T. Ng, Vladimir Tucakov. Distance-based outliers: algorithms and applications. The VLDB Journal — The International Journal on Very Large Data Bases, volume 8, 2000.
[6] Jue Wang, Michael F. Cohen. Image and video matting: a survey. Foundations and TrendsR in Computer Graphics and Vision, 2007.
[7] Y.Y. Chuang, B. Curless, D. Salesin, R. Szeliski. A Bayesian Approach to Digital Matting. Computer Vision and Pattern Recognition, 2001.
[8] J. Sun, J. Jia, C.K. Tang, H.Y. Shum. Poisson Matting. SIGGRAPH, 2004.
[9] A. Levin, D. Lischinski, Y. Weiss. A Closed Form Solution to Natural Image Matting. Computer Vision and Pattern Recognition, 2006.
[10] J. Wang, M. Cohen. Optimized Color Sampling for Robust Matting. Computer Vision and Pattern Recognition, 2007.
[11] Xue Bai, Guillermo Sapiro. A geodesic framework for fast interactive image and video segmentation and matting. International Conference on Computer Vision, 2007.
[12] Christoph Rhemann, Carsten Rother, Jue Wang, Margrit Gelautz, Pushmeet Kohli, Pamela Rott. A Perceptually Motivated Online Benchmark for Image Matting. Computer Vision and Pattern Recognition, 2009.
[13] U. Lipowezky, O. Mamo, A. Cohen. Using integrated color and texture features for automatic hair detection. In Convention of Electrical and Electronics Engineers in Israel, 2008.
[14] C. Rousset, P. Y. Coulon. Frequential and color analysis for hair mask segmentation. In International Conference on Image Processing, 2008.
[15] P. Julian and C. Dehais and Lauze and V. Charvillat and A. Bartoli and A. Choukroun. Automatic Hair Detection in the Wild. International Conference on Pattern Recognition, 2010.