Journal of Electrical and Computer Engineering

Research Article

No-Reference Stereoscopic IQA Approach: From Nonlinear Effect to Parallax Compensation

Table 4

PLCC, SRCC, KRCC, AAE, and RMSE values (after nonlinear regression) of testing and training groups in and NOSPDM₁–NOSPDM₁₈.


Metrics	PLCC	SRCC	KRCC	AAE	RMSE

(testing)	0.9350	—	—	0.3500	0.4210
(training)	0.9660	—	—	0.2920	0.3670
	0.8934	0.8897	0.7066	0.4136	0.5234
	0.8943	0.8912	0.7097	0.4120	0.5212
	0.9350	0.9312	0.7666	0.3303	0.4133
	0.9360	0.9325	0.7699	0.3269	0.4100
	0.9318	0.9266	0.7609	0.3358	0.4228
	0.9288	0.9231	0.7556	0.3416	0.4316
	0.9493	0.9433	0.7926	0.2851	0.3664
	0.9665	0.9600	0.8284	0.2330	0.2992
	0.9705	0.9652	0.8416	0.2146	0.2807
	0.9502	0.9444	0.7951	0.2814	0.3630
	0.9673	0.9611	0.8313	0.2289	0.2955
	0.9711	0.9660	0.8442	0.2114	0.2782
	0.9470	0.9400	0.7878	0.2896	0.3741
	0.9638	0.9565	0.8223	0.2402	0.3106
	0.9676	0.9613	0.8340	0.2238	0.2942
	0.9445	0.9373	0.7837	0.2958	0.3827
	0.9610	0.9534	0.8162	0.2476	0.3220
	0.9647	0.9581	0.8278	0.2344	0.3067