Fast pose invariant face recognition using super coupled multiresolution Markov Random Fields on a GPU

Fast pose invariant face recognition using super coupled multiresolution Markov Random Fields on a GPU
复制标题

在 GPU 上使用超耦合多分辨率马尔可夫随机场进行快速姿势不变人脸识别

DOI:
10.1016/j.patrec.2014.05.017
复制
发表时间:
2014
影响因子:
5.1
通讯作者:
Rahimzadeh Arashloo S
Rahimzadeh Arashloo S
中科院分区:
计算机科学3区
文献类型:
--
作者:
Rahimzadeh Arashloo S

文献摘要

参考文献

被引文献

相似文献

我们讨论的问题,姿态不变的人脸识别使用马尔可夫随机场(MRF)模型。在早期研究中,MRF图像与图像匹配已被证明非常有前途(Arashloo和Kittler,2011)[4]。Arashloo等人(2011)[6]通过Petrou等人(1998)[37,11]倡导的超级耦合变换链接的多分辨率MRF解决了其苛刻的计算复杂性。在本文中,我们受益于菊花描述符的人脸图像表示的图像匹配。最重要的是,我们设计了所提出的多分辨率MRF匹配过程的创新GPU实现。所实现的显著加速(25倍)具有多个好处:它使MRF方法成为一个实用的建议。它有利于广泛的经验优化和评估研究。后者进行基准数据库,包括具有挑战性的标记在野生(LFW)数据库中的脸显示了所提出的方法,它始终实现了国家的最先进的性能在标准的基准测试的杰出潜力。实验研究还表明,超耦合多分辨率MRF提供了一个5倍以上的计算速度超过使用GPU实现实现的速度。
We discuss the problem of pose invariant face recognition using a Markov Random Field (MRF) model. MRF image to image matching has been shown to be very promising in earlier studies (Arashloo and Kittler, 2011) [4]. Its demanding computational complexity has been addressed in Arashloo et al. (2011) [6] by means of multiresolution MRFs linked by the super coupling transform advocated by Petrou et al. (1998) [37, 11]. In this paper, we benefit from the daisy descriptor for face image representation in image matching. Most importantly, we design an innovative GPU implementation of the proposed multiresolution MRF matching process. The significant speed up achieved (factor of 25) has multiple benefits: It makes the MRF approach a practical proposition. It facilitates extensive empirical optimisation and evaluation studies. The latter conducted on benchmarking databases, including the challenging labelled faces in the wild (LFW) database show the outstanding potential of the proposed method, which consistently achieves state-of-the-art performance in standard benchmarking tests. The experimental studies also show that the super coupled multiresolution MRFs deliver a computational speed up by a factor of 5 over and above the speed up achieved using the GPU implementation.
DOI: 10.1016/j.cviu.2010.12.006
发表时间: 2011-07
期刊: Comput. Vis. Image Underst.
影响因子: --
作者:
Shervin Rahimzadeh Arashloo;J. Kittler;W. Christmas
通讯作者: Shervin Rahimzadeh Arashloo;J. Kittler;W. Christmas
DOI: 10.5555/2503308.2188386
发表时间: 2012
期刊: J. Mach. Learn. Res.
影响因子: --
作者:
Yiming Ying;Peng Li
通讯作者: Yiming Ying;Peng Li
使用超级耦合方法的非线性运动估计
DOI: 10.1109/34.682185
发表时间: 1998
期刊: IEEE Trans. Pattern Anal. Mach. Intell.
影响因子: --
作者:
M. Bober;M. Petrou;J. Kittler
通讯作者: J. Kittler
DOI: 10.1007/11744023_32
发表时间: 2006-01-01
期刊: COMPUTER VISION - ECCV 2006 , PT 1, PROCEEDINGS
影响因子: --
作者:
Bay, Herbert;Tuytelaars, Tinne;Van Gool, Luc
通讯作者: Van Gool, Luc
DOI: 10.1109/icpr.1994.577123
发表时间: 1994
期刊: Proceedings of the 12th IAPR International Conference on Pattern Recognition, Vol. 2 - Conference B: Computer Vision & Image Processing. (Cat. No.94CH3440-5)
影响因子: --
作者:
G. Nicholls;M. Petrou
通讯作者: M. Petrou