An efficient compression and reconstruction method of face image for low rate net

  • Authors:
  • Xing Li;JianHuang Lai;ZhiBin Zhang

  • Affiliations:
  • Center of Computer Vision, Sun Yat-sen University, Guangzhou, China;Center of Computer Vision, Sun Yat-sen University, Guangzhou, China;Center of Computer Vision, Sun Yat-sen University, Guangzhou, China

  • Venue:
  • SINOBIOMETRICS'04 Proceedings of the 5th Chinese conference on Advances in Biometric Person Authentication
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

A generic 2D face image coding method used for face recognition or expression reconstruction in the low rate net is introduced in this article By this method the author reconstructs facial texture images with partial areas of reference face pictures by least-square minimization (LSM) with a high compression ratio and a good PSNR value Our experimental results show that reconstructed faces are very natural and plausible like real photos This technique has improved outline shape control ability of PCA method, and at the same time, it can maintain details adjust ability of PCA method.