A fast video super resolution for facial image

作者：

Highlights：

• Multi-frame super-resolution for facial images is addressed.

• Key points from faces are found.

• Triangular patches are determined from pixels of the reference frame.

• Energy function is minimized while considering both appearance and pixel displacement.

• The proposed method has low computational complexity.

摘要

•Multi-frame super-resolution for facial images is addressed.•Key points from faces are found.•Triangular patches are determined from pixels of the reference frame.•Energy function is minimized while considering both appearance and pixel displacement.•The proposed method has low computational complexity.

论文关键词：Video super resolution,Dense registration,Facial image

论文评审过程：Received 24 February 2018, Revised 24 October 2018, Accepted 24 October 2018, Available online 30 October 2018, Version of Record 12 November 2018.

论文官网地址：https://doi.org/10.1016/j.image.2018.10.008