A Simple Quality Assessment Index for Stereoscopic Images Based on 3D Gradient Magnitude

Table 1

Performance of the proposed method and the other seven schemes in terms of PLCC, SRCC, and RMSE on the four databases (the cases in bold: the best performance).

IQA model	NBU (312 images)			LIVE I (365 images)			LIVE II-S (120 images)			LIVE II-A (240 images)
	PLCC	SRCC	RMSE	PLCC	SRCC	RMSE	PLCC	SRCC	RMSE	PLCC	SRCC	RMSE
PSNR	0.8255	0.8519	9.6960	0.8354	0.8339	9.0117	0.7651	0.7768	8.0389	0.6659	0.6752	7.5610
SSIM	0.8347	0.8575	9.4582	0.8887	0.8873	7.5155	0.7765	0.7488	7.8656	0.7676	0.7388	6.4949
MS-SSIM	0.8510	0.9295	9.0213	0.9287	0.9224	6.0771	0.8824	0.9077	5.8782	0.7329	0.7093	6.8947
Benoit [13]	0.7838	0.8118	10.6675	0.8786	0.8852	7.8281	0.8312	0.8412	6.9411	0.7622	0.7342	6.5613
You [14]	0.8205	0.8246	9.8196	0.9172	0.9248	6.5328	0.9190	0.9491	4.9206	0.7469	0.7184	6.7388
Bensalma [17]	0.9378	0.9381	5.9615	0.8902	0.8746	7.4683	0.8539	0.8418	6.4956	0.7663	0.7210	6.5111
Chen [18]	0.9388	0.9374	5.9153	0.9220	0.9078	6.3474	0.8511	0.8624	6.6044	0.6317	0.6301	7.9343
Shao [20]	0.9266	0.9271	6.4597	0.9270	0.9217	6.1497	0.9286	0.9153	4.6323	0.6098	0.6300	8.0329
Proposed	0.9240	0.9331	6.5711	0.9213	0.9158	6.3748	0.9515	0.9443	3.8411	0.7277	0.6951	6.9520