Improved structure tensor for fine-grained texture inpainting
作者：
Highlights：
• An effective inpainting method for restoring damaged texture images is presented.
• Our method extends structure-tensor-based partial differential equations (PDEs).
• The algorithm is based on a fractional structure tensor with double-sampling (DFST).
• Non-locally regularized DFST (NDFST) was then inserted into an anisotropic PDE.
• The resulting method recovers non-homogeneous structures and fine-grained textures.
摘要
•An effective inpainting method for restoring damaged texture images is presented.•Our method extends structure-tensor-based partial differential equations (PDEs).•The algorithm is based on a fractional structure tensor with double-sampling (DFST).•Non-locally regularized DFST (NDFST) was then inserted into an anisotropic PDE.•The resulting method recovers non-homogeneous structures and fine-grained textures.
论文关键词：Texture,Inpainting,Structure tensor,Non-local,Fractional-order
论文评审过程：Received 10 August 2017, Revised 26 January 2018, Accepted 5 February 2018, Available online 12 February 2018, Version of Record 12 March 2019.
论文官网地址：https://doi.org/10.1016/j.image.2018.02.006
