Image inpainting based on sparse representations with a perceptual metric

EURASIP Journal on Advances in Signal Processing

Table 2 Performance comparison (SSIM) of the proposed method and existing methods

Image number	[31]	[34]	[35]	[24]	[30]	[41]	Proposed method
Image 1 (Figure4)	0.6355	0.6090	0.6411	0.6822	0.6773	0.7145	0.7811
Image 2 (Figure5)	0.5130	0.5154	0.5999	0.5077	0.5277	0.5308	0.6454
Image 3 (Figure6)	0.6248	0.6051	0.6538	0.7318	0.7246	0.7569	0.8156
Image 4 (Figure10, first column)	0.5833	0.5762	0.6373	0.6563	0.6708	0.7036	0.7806
Image 5 (Figure10, second column)	0.6419	0.6424	0.6774	0.7298	0.7196	0.7410	0.8049
Image 6 (Figure10, third column)	0.6460	0.6458	0.7346	0.6750	0.6933	0.6756	0.7619
Image 7 (Figure10, fourth column)	0.6711	0.6766	0.7134	0.7478	0.7521	0.7402	0.7722
Image 8 (Figure11, first column)	0.5871	0.5522	0.6282	0.6561	0.6394	0.6840	0.6940
Image 9 (Figure11, second column)	0.6501	0.6599	0.7645	0.6852	0.6799	0.6700	0.8072
Image 10 (Figure11, third column)	0.6240	0.6295	0.7185	0.6992	0.6980	0.7060	0.7681
Image 11 (Figure11, fourth column)	0.6864	0.7069	0.7685	0.7155	0.7108	0.7352	0.8039