3D Data Denoising via Nonlocal Means Filter by Using Parallel GPU Strategies

Table 8

Partial unrolling algorithms for different video formats. Observe that the execution times include the time needed to copy data from CPU to GPU and from GPU to CPU.

Video size	MB	Execution time
		(11^3,3^3)	(21^3,3^3)	(11^3,5^3)	(21^3,5^3)
PAL ×50	79.1	20.8	145	72.9	508
PAL ×100	158	41.6	291	146	1021
PAL ×200	316	83.7	586	294	2045
PAL ×500	791	212	1500	744	5162
HD ×50	148	39	271	136	953
HD ×100	297	78.1	545	274	1914
HD ×200	593	157	1104	551	3848
HD ×500	1483	398	2812	1395	9678
FULL-HD ×50	396	97.3	678	342	2386
FULL-HD ×100	791	195	1365	686	4788
FULL-HD ×200	1582	392	2754	1375	9601
FULL-HD ×500	3955	995	7018	3483	24149