In this paper we present a trained diffusion model for image inpainting based on the structural similarity measure. The proposed diffusion model uses several parametrized linear filters and influence functions. Those parameters are learned in a loss based approach, where we first perform a greedy training before conducting a joint training to further improve the inpainting performance. We provide a detailed comparison to state-of-the-art inpainting algorithms based on the TUM-image inpainting database. The experimental results show that the proposed diffusion model is efficient and achieves superior performance. Moreover, we also demonstrate that the proposed method has a texture preserving property, that makes it stand out from previous PDE based methods.
|Title of host publication||Pattern Recognition|
|Subtitle of host publication||37th German Conference, GCPR 2015, Aachen, Germany, October 7-10, 2015, Proceedings|
|Publisher||Springer International Publishing AG|
|Publication status||Accepted/In press - 2015|
Fields of Expertise
- Information, Communication & Computing
Yu, W., Heber, S., & Pock, T. (Accepted/In press). Learning Reaction-Diffusion Models for Image Inpainting. In Pattern Recognition: 37th German Conference, GCPR 2015, Aachen, Germany, October 7-10, 2015, Proceedings (Vol. 9358, pp. 356-367). Springer International Publishing AG . https://doi.org/10.1007/978-3-319-24947-6_29