Low Resolution Face Recognition using Super-resolution

Hyunduk Kim, Sang-Heon Lee, Myoung-Kyu Sohn, Dong-Ju Kim, Byungmin Kim


With the increasing demand of surveillance camerabased applications, the low resolution face recognition algorithms need in many face application system. Recognition performance will be dramatically degraded under the low resolution. In order to overcome this problem, face super-resolution methods can be employed to enhance the resolution of the images. In this paper low resolution face recognition algorithm using new superresolution is introduced in which the resolution of the face image to be recognized is 166. Experiments of super-resolution image reconstruction were performed with the L2 error norm for datafidelity and proposed weighted gradient constraint were used as regularization terms. Additionally, the eigenface and fisherface were employed for face recognition. Experimental results show that the proposed SR algorithm outperforms the existing algorithms in public face databases.


super-resolution; regularization technique; low resolution face recognition; bilateral filter

Full Text:



  • There are currently no refbacks.