研究生: |
張仁豪 Jen-Hao Chang |
---|---|
論文名稱: |
基於單張相片之肖像重建 Face Reconstruction From Single Photo |
指導教授: |
姚智原
Chih-Yuan Yao |
口試委員: |
賴祐吉
Yu-Chi Lai 朱宏國 Hung-Kuo Chu 王昱舜 Yu-Shuen Wang |
學位類別: |
碩士 Master |
系所名稱: |
電資學院 - 資訊工程系 Department of Computer Science and Information Engineering |
論文出版年: | 2015 |
畢業學年度: | 103 |
語文別: | 中文 |
論文頁數: | 46 |
中文關鍵詞: | 人臉特徵 、3D變形 、人臉貼圖 |
外文關鍵詞: | Facial features, 3D deformation, face texture |
相關次數: | 點閱:161 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
近來手機APP的市場裡,已出現幾款利用單張相片來進行人臉建模的APP,其中最顯而易見的缺點為其結果與原人臉的相似度並沒有想像中的高,因為它們最常利用的方法是將2D人臉圖片直接貼上固定模板的3D模型上。因此,從一開始就能從模板的3D模型跟原始人臉圖片的外型相似度看出最終建模結果的相似程度,而本論文則將提出一個讓3D模型更貼近原始人臉的一種全新方式。
Recently on the app store, a few apps that use a single photo to model human faces have been released. The most obvious flaw of the apps is the results are not as similar to the original face as expected, since the most common method they use is to apply a 2D face image directly to a predefined 3D face template model. Therefore, the similarity between the final result and the original face photo can be observed from the similarity between the 3D template model and the original face photo. In this paper, we propose a novel method for creating a 3D model that has more likeness to the original human face.
[1] Mark Pauly Alexandru-Eugen Ichim, So en Bouaziz. Dynamic 3d avatar creation from hand-held video input. ACM Transactions on Graphics (Proceedings of SIGGRAPH), 2015.
[2] J. Cho, S. Mirzaei, J. Oberg, and R. Kastner. Fpga-based face detection system using haar classi ers. Proceedings of the ACM/SIGDA international symposium on Field programmable gate arrays, pages 103-112, 2009.
[3] S. Lucey J. Saragih and J. Cohn. Deformable model tting by regularized landmark mean-shift. International Journal of Computer Vision (IJCV), 91(2):pp 200-215, 2010.
[4] Alec Jacobson, Ilya Baran, Jovan Popovi c, and Olga Sorkine. Bounded biharmonic weights for real-time deformation. ACM Transactions on Graphics
(proceedings of ACM SIGGRAPH), 30(4):78:1-78:8, 2011.
[5] Alec's Web Log. Areas of quadrilaterals within a triangle determined by its circumcenter (or perpendicular bisectors), using matlab, part 2. http://www.alecjacobson.com/weblog/?p=874. Accessed: 2015-06-30.
[6] R. Schmidt, C. Grimm, and B. Wyvill. Interactive decal compositing with
discrete exponential maps. Proc. ACM SIGGRAPH '06, 25:605-613, 2006.
[7] O. Sorkine, D. Cohen-Or, Y. Lipman, M. Alexa, C. Rossl, and H.-P. eidel. Laplacian surface editing. In Proc. Eurographics/ACM SIGGRAPH Symposium '04, pages 179-188, 2004.
[8] Kenshi Takayama, Daniele Panozzo, Alexander Sorkine-Hornung, and Olga
Sorkine-Hornung. Sketch-based generation and editing of quad meshes. ACM
Transactions on Graphics (proceedings of ACM SIGGRAPH), 32(4), 2013.
[9] P. Viola and M. Jones. Rapid object detection using a boosted cascade of simple features. 1:511-518, 2001.
[10] Paul Viola and Michael J. Jones. Robust real-time face detection. International Journal of Computer Vision (IJCV), 57(2):pp 137-154, 2004.
[11] Y. Wu, Z. Wang, and Q. Ji. Facial feature tracking under varying facial expressions and face poses based on restricted boltzmann machines. In CVPR, pages 3452-3459. IEEE, 2013.
[12] Xiaoguang Yan. Constrained Local Model for Face Alignment, a Tutorial, 2011.
[13] S. Zhang, A. Nealen, and D. Metaxas. Skeleton based as-rigid-as-possible volume modeling. EUROGRAPHICS, 2010.