Table 2.

Results for a subset of 15 images from [16] (images selected as detailed in Section 4). The underlined images are depicted in Figure 5 and Figure 6 for qualitative evaluation. Bold values correspond to the best performance in the given similarity measurement; #t corresponds to the number of times a given algorithm obtained the best performance. The last section (HEP) contains the results from the survey; each value corresponds to the number of users that selected this result as the best one. ”SAME” is used when the output from both architectures was the same for a given user.

Image #	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
Method	PSNR															#t
X	26.16	11.50	25.15	16.29	16.49	13.41	15.20	18.23	12.66	15.32	16.29	16.34	25.80	17.24	15.41	-
[20]	24.99	21.38	27.04	18.76	20.89	19.84	19.76	22.18	18.81	19.77	15.31	17.86	22.07	24.63	17.30	5
[21]	18.70	12.68	22.70	15.17	12.69	13.50	15.88	16.60	12.64	18.90	16.57	17.78	19.24	13.87	14.06	0
SRCNN [30]	28.04	22.82	19.88	14.85	16.14	18.01	15.48	18.43	17.33	18.57	17.79	18.42	22.92	19.80	12.93	0
ENDENet	33.51	17.96	26.07	20.31	23.10	18.28	21.59	23.27	21.10	19.55	18.84	19.24	26.79	26.28	15.48	5
CDNet	34.44	15.19	26.00	20.04	23.00	18.09	22.25	23.39	21.74	19.62	18.63	17.88	27.20	26.13	15.32	5
Method	SSIM
X	0.91	0.45	0.59	0.60	0.70	0.35	0.62	0.59	0.53	0.41	0.49	0.42	0.87	0.74	0.64	-
[20]	0.79	0.70	0.69	0.49	0.63	0.51	0.49	0.59	0.52	0.65	0.56	0.54	0.79	0.80	0.64	4
[21]	0.36	0.24	0.36	0.34	0.19	0.22	0.36	0.31	0.17	0.39	0.20	0.24	0.36	0.34	0.31	0
SRCNN [30]	0.87	0.67	0.52	0.36	0.49	0.45	0.37	0.50	0.54	0.60	0.58	0.44	0.72	0.70	0.48	0
ENDENet	0.96	0.69	0.62	0.60	0.71	0.49	0.67	0.73	0.58	0.62	0.70	0.55	0.85	0.84	0.65	4
CDNet	0.96	0.63	0.61	0.63	0.73	0.39	0.69	0.73	0.63	0.59	0.68	0.52	0.87	0.85	0.67	9
Method	AE
X	1.40	5.06	8.82	3.37	3.25	9.55	6.27	8.78	4.77	5.54	4.80	7.81	1.68	4.07	4.28	-
[20]	3.55	5.28	8.12	4.24	3.45	6.22	8.50	8.30	7.63	3.45	4.02	5.89	2.72	3.09	3.83	2
[21]	1.54	5.51	9.18	3.69	3.63	5.49	7.95	7.79	7.96	3.96	4.50	6.67	2.74	4.34	3.71	1
SRCNN [30]	1.37	4.91	12.51	5.47	5.41	6.15	9.31	11.61	5.52	3.63	3.47	5.81	3.13	5.63	6.74	2
ENDENet	0.76	3.76	8.79	3.66	3.38	6.80	5.80	6.04	9.32	3.46	3.03	5.89	2.05	2.53	4.16	6
CDNet	0.75	4.13	9.14	4.02	3.76	8.07	5.43	6.30	6.56	4.08	3.56	6.19	2.02	2.71	3.78	4
Method	$\Delta E_{00}$
X	84.52	7.70	95.34	26.44	26.64	13.38	43.92	64.98	48.51	26.85	26.40	18.52	91.64	72.96	31.71	-
[20]	99.43	90.37	93.14	37.54	60.69	65.58	71.02	80.80	75.99	48.36	15.87	16.70	58.91	88.38	31.72	3
[21]	30.06	11.40	90.91	23.70	14.29	21.18	53.74	59.02	22.99	37.30	11.39	16.70	46.96	28.02	26.37	0
SRCNN [30]	95.10	93.30	97.52	8.82	13.20	55.92	29.71	58.47	62.46	45.82	22.11	16.87	82.20	71.33	22.61	0
ENDENet	99.65	61.14	93.31	42.56	70.24	61.78	81.76	90.36	77.89	56.63	52.88	27.27	92.48	91.84	26.40	6
CDNet	99.76	36.32	92.99	43.71	68.14	46.65	84.46	87.82	85.41	50.55	37.81	19.27	93.72	92.11	26.56	6
Method	HEP
SAME	32	8	5	9	15	8	22	16	11	13	5	6	28	29	16	3
ENDENet	13	36	40	22	5	10	4	15	4	4	37	12	11	4	12	4
CDNet	5	6	5	19	30	32	24	19	35	33	8	32	11	17	22	8