Variational single image interpolation with time-varying regularization

作者：

Highlights：

•

• A variational model with time-varying regularization for single image interpolation.

• Strongly competitive interpolation performance compared to state-of-the-art methods.

• Highly efficient, about 2 to 4 orders of magnitude faster than SOTAs.

摘要

•A variational model with time-varying regularization for single image interpolation.•Strongly competitive interpolation performance compared to state-of-the-art methods.•Highly efficient, about 2 to 4 orders of magnitude faster than SOTAs.

论文关键词：Image interpolation,Nonlinear diffusion,Regularization,Loss specific training

论文评审过程：Received 7 October 2016, Revised 3 August 2017, Accepted 4 August 2017, Available online 23 August 2017, Version of Record 1 September 2017.

论文官网地址：https://doi.org/10.1016/j.image.2017.08.005