notesum.ai

Published at November 29

Subjective and Objective Quality Assessment Methods of Stereoscopic Videos with Visibility Affecting Distortions

cs.MM

cs.CV

Released Date: November 29, 2024

Authors: Sria Biswas¹, Balasubramanyam Appina², Priyanka Kokil¹, Sumohana S Channappayya³

Aff.: ¹Institution 1; ²Institution 2; ³Institution 3

Arxiv: http://arxiv.org/pdf/2411.19522v1

Refer to caption

Model Type	Algorithm	Symmetric			Asymmetric			Overall
Model Type	Algorithm	LCC	SROCC	RMSE	LCC	SROCC	RMSE	LCC	SROCC	RMSE
2D FR IQA	SSIM [78]	0.425	0.399	14.696	0.408	0.388	14.706	0.417	0.391	14.747
	MS-SSIM [79]	0.458	0.407	14.600	0.433	0.371	14.576	0.439	0.402	14.637
	VIF [80]	0.568	0.548	13.285	0.480	0.430	14.126	0.498	0.435	13.523
2D NR IQA	BRISQUE [81]	0.251	0.184	15.628	0.185	0.219	16.093	0.178	0.199	16.209
	NIQE [82]	0.421	0.352	14.379	0.327	0.294	15.693	0.380	0.356	15.649
	PIQUE [83]	0.231	0.209	15.739	0.216	0.208	15.765	0.219	0.206	15.090
	ILNIQE [84]	0.469	0.459	14.255	0.323	0.241	15.406	0.374	0.319	14.961
2D FR VQA	STMAD [85]	0.754	0.686	8.810	0.621	0.589	11.171	0.677	0.628	10.424
2D FR VQA	SpEED [86]	0.757	0.700	8.809	0.715	0.696	9.310	0.749	0.700	9.558
2D NR VQA	VIIDEO [87]	0.262	0.234	15.582	0.286	0.271	15.758	0.272	0.255	15.895
2D NR VQA	NSTSS [88]	0.418	0.289	14.766	0.396	0.278	14.829	0.395	0.283	14.830
3D FR IQA	STRIQE [69]	0.601	0.602	12.894	0.566	0.534	11.814	0.577	0.569	11.773
3D FR IQA	MJ3DQA [39]	0.654	0.624	10.955	0.589	0.491	11.958	0.617	0.581	10.823
3D NR IQA	MO-NIQE [42]	0.229	0.200	15.973	0.165	0.136	16.319	0.213	0.193	15.085
3D FR VQA	$\text{DeMo}_{3D}$ [49]	0.753	0.665	8.811	0.625	0.610	11.169	0.680	0.641	10.422
3D NR VQA	VQUEMODES [50]	0.873	0.834	3.166	0.846	0.833	5.398	0.845	0.849	5.239
	$\text{MoDi}_{3D}$ [28]	0.486	0.447	14.148	0.336	0.302	15.411	0.407	0.632	14.716
	CBSE	0.767	0.749	8.750	0.663	0.643	10.610	0.717	0.685	9.805