当前位置>主页 > 期刊在线 > 信息技术 >

信息技术

基于稀疏表示与人脸结构的人脸幻构技术研究
黎兆文,胡晓
(广州大学机械与电气工程学院,广东 广州 510006)
摘要点击次数:163    

摘  要:虽然基于稀疏表示的方法重构人脸图像有着样本库需求小的优点,同时对于平滑区域的恢复也有很好的效果,但是人脸成分以及轮廓边缘细节仍然较为模糊。为了解决这一问题,本文提出了结合稀疏表示的梯度估计边缘优化方法,该方法利用样本库中高分辨率的人脸成分以及边缘梯度统计空间对低分辨率输入人脸进行细节恢复和边缘锐化。实验结果表明,该方法对人脸图像结构的细节恢复有较为理想的表现效果。


关键词: 稀疏表示;梯度估计;人脸结构;人脸幻构


作者介绍:

黎兆文(1991.08-),男,汉族,广东人,硕士。研究方向:人脸图像处理。通信作者:胡晓(1969-),男,湖南人,教授,硕士。研究方向:智能信号处理;人脸检测和识别;医学信号处理。


中图分类号:TP391.41     文献标识码:A 文章编号:2096-4706(2018)03-0000-06

Face Animation Technology Based on SparseRepresentation and Face Structure

LI Zhaowen,HU Xiao

(School of Mechanical and ElectricalEngineering,Guangzhou University,Guangzhou  510006,China)

Abstract:Although the reconstruction of face images based on sparserepresentation has the advantages of small sample base and good effect on therecovery of the smooth region,the face composition andthe outline edge details are still relatively vague. In order to solve this problem,this paper proposes a gradient estimation edge optimization methodcombined with sparse representation,which uses highresolution face components and edge gradient statistical space in sampledatabase to restore and sharpen the edges of low resolution input faces. Experimentalresults show that this method has a satisfactory effect on detail restorationof face image structure.

Keywords:sparse representation;gradient estimation;face structure;face magic


参考文献:

   [1] S.Baker,T.Kanade.“Hallucinatingfaces,”in Fourth IEEEInternational Conference on Automatic Face and Gesture Recognition,2000.Proceedings,pp.83–88,IEEE,2000.

   [2] C.Liu,H.-Y.Shum,and C.-S.Zhang.“A two-step approach to hallucinating faces:global parametric model and local nonparametric model,”in Proceedings of the 2001 IEEEComputer Society Conference on Computer Vision and Pattern Recognition,2001.vol.1,pp.I–192.

   [3] W.T.Freeman,T.R.Jonesand E.C.Pasztor,“Examplebased super-resolution,” Computer Graphics andApplications,IEEE,,vol.22,no.2,pp.56–65,2002.

   [4] YANG J C,WANG ZW,LIN Z,et al.Coupled dictionary learning for image super-resolution [J].IEEETransactions on Image Processing,2012,21(8):3467-3478.

   [5]WRIGHT Y J,HUANGTS. Image super-resolution via sparse representation[J]. IEEE Transactions onImage Processing,2010,19(11):2861-2873.

   [6] ZEYDE R,ELAD M,PROTTER M.On single imagescale-up usingsparse representations [C].//International Conference on Curves and Surfaces.Springer-Verlag,2010:711-730.

   [7] AI N,PENG J Y,ZHU X,et al.SISR viatraineddouble sparsity dictionaries [J].Multimedia ToolsandApplications,2015,74(6):1997-2007.

   [8] W.T.Freeman,E.C.Pasztor,O.T.Carmichael,“Learning lowlevel vision,”Int. J.Comput. Vis,vol.40,no.1,pp.25–47,2000

   [9] H.Chang,D.-Y.Yeung,Y.Xiong.“Super-resolution throughneighbor embedding,” in Proc. IEEE Conf. Comput.Vis.PatternClass.2004,vol.1 pp

   [10] S.Baker and T.Kanade,“Limitson super-resolution and how to breakthem,”IEEE Trans.Pattern Anal.Mach. Intell.vol.24,no.9,pp.1167–1183,Sep.2002.

   [11] LiY,Liu J,Yang W,et al.Multi-pose face hallucination via neighbor embedding forfacial components [C].//IEEE International Conference on Image Processing.IEEE,2015:4037-4041.

   [12] C.-Y.Yang,S.Liu,and M.-H.Yang..“Structured face hallucination,”in Proc.IEEE Int’lConf.Computer Vision and Pattern Recognition,2013.

   [13] M.F.Tappen and C.Liu.A Bayesianapproach to alignmentbased image hallucination.In ECCV,2012.

   [14] C.Liu,H.-Y.Shum,and W.T.Freeman.Face hallucination:Theoryand practice.IJCV,75(1):115–134,2007.

   [15] X.Ma,J.Zhang,and C.Qi.Hallucinating face by positionpatch.PR,43(6):2224–2236,2010.

   [16] R.Fattal.Image upsampling viaimposed edge statistics.In SIGGRAPH,2007.

   [17] J.Sun,J.Sun,Z.Xu and H.-Y.Shum.Image superresolution using gradient profileprior.In CVPR,2008.

[18]GROSS R,MATTHEWSI,COHN J,et al.Multi-pie[C].2008-06-15,S.l.:s.n.,2008:1-8.

   [19] M.Irani and S.Peleg.Improvingresolution by image registration [J].CVGIP,53(3):231–239,1991.