Image inpainting based on low-rank and joint-sparse matrix recovery

Image inpainting is a classical inverse problem of image science and has many applications. In the previous works, most of the variational inpainting methods can be considered as special cases of the restoration model where the linear operator is just the project to the known indexes. In this reported work, the variational inpainting model is established from the view of image decomposition. Then the unknown component can be recovered by the known component under the low-rank and joint-sparse constraints. Numerical experiments demonstrate that the proposed algorithm outperforms most of the current state-of-the-art methods with respect to the peak-signal-to-noise ratio value.