Video super-resolution using an adaptive superpixel-guided auto-regressive model

作者：

Highlights：

• An automatic key-frame selection method is proposed.

• An adaptive superpixel-guided auto-regressive (AR) model is proposed.

• Our method has the best visual quality and the highest PSNR.

• Our method requires less computation and is suitable for practical applications.

摘要

Highlights•An automatic key-frame selection method is proposed.•An adaptive superpixel-guided auto-regressive (AR) model is proposed.•Our method has the best visual quality and the highest PSNR.•Our method requires less computation and is suitable for practical applications.

论文关键词：Video super-resolution,Superpixel,Auto-regressive model,Spatio-temporal correlation

论文评审过程：Received 28 April 2014, Revised 27 April 2015, Accepted 7 August 2015, Available online 18 August 2015, Version of Record 27 November 2015.

论文官网地址：https://doi.org/10.1016/j.patcog.2015.08.008