Image Inpainting Based on Probabilistic Structure Estimation
A novel inpainting method based on probabilistic structure estimation has been developed. The method consists of two steps. First, an initial image, which captures rough structure and colors in the missing region, is estimated. This image is generated by probabilistically interpolating the gradient inside the missing region, and then by flooding the colors on the boundary into the missing region using Markov Random Field. Second, by locally replacing the missing region with local patches similar to both the adjacent patches and the initial image, the inpainted image is synthesized. Since the patch replacement process is guided by the initial image, the inpainted image is guaranteed to preserve the underlying structure. This also enables patches to be replaced in a greedy manner, i.e. without optimization. Experiments show the proposed method outperforms previous methods in terms of both subjective image quality and computational speed.
KeywordsMarkov Random Field Initial Image Missing Region Subjective Image Quality Greedy Method
Unable to display preview. Download preview PDF.
- 1.Barnes, C., Shechtman, E., Finkelstein, A., Goldman, D.B.: Patchmatch: A randomized correspondence algorithm for structural image editing. In: Proc. of ACM SIGGRAPH, vol. 29 (2009)Google Scholar
- 2.Bertalmio, M., Sapiro, G., Caselles, V., Ballester, C.: Image inpainting. In: Proc. of ACM SIGGRAPH, pp. 417–424 (2000)Google Scholar
- 4.Chan, F.T., Shen, J.: Non-texture inpainting by curvature-driven diffusions. J. of VCIR 12, 436–449 (2001)Google Scholar
- 5.Chen, Y., Luan, Y., Li, H., Au, C.O.: Sketch-guided texture-based image inpainting. In: Proc. of ICIP, pp. 1997–2000 (2006)Google Scholar
- 6.Criminisi, A., Pérez, P., Toyama, K.: Region filling and object removal by exemplar-based image inpainting. IEEE Trans. on IP 13, 1200–1212 (2004)Google Scholar
- 7.Harrison, P.: A Non-hierarchical procedure for re-synthesis of complex texture. In: Proc. of WSCG, pp. 190–197 (2001)Google Scholar
- 8.Jia, J., Tang, K.C.: Image repairing: robust image synthesis by adaptive ND tensor voting. In: Proc. of CVPR, pp. 643–650 (2003)Google Scholar
- 9.Kawai, N., Sato, T., Yokoya, N.: Image inpainting considering brightness change and spatial locality of textures. In: Proc. of VISAPP, vol. 1, pp. 66–73 (2008)Google Scholar
- 11.Li, R.B., Qi, Y., Shen, K.X.: An image inpainting method. In: Conf. on CAD and Computer Graphics, pp. 531–536 (2005)Google Scholar
- 12.Pritch, Y., Kav-Venaki, E., Peleg, S.: Shift-map image editing. In: Proc. of ICCV, pp. 151–158 (2009)Google Scholar
- 13.Sun, J., Yuan, L., Jia, J., Shum, Y.H.: Image completion with structure propagation. In: Proc. of ACM SIGGRAPH, pp. 861–868 (2005)Google Scholar