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

Table 6

Full unrolling algorithm on a single GPU unit. Execution times and speed-up values for a 3D dataset of normally distributed random numbers (size = 512×512×128) for several window configurations.

(|𝑉𝑖|,|𝑗𝐵𝑖|)	Execution time/speed-up
	1 GPU unit				CPU
	(16, 16, 1)	(128, 1, 1)	(256, 1, 1)	(512, 1, 1)
(11^3,3^3)	71.4/39.4	70/40.2	70.1/40.1	73.9/38.1	2814
(21^3,3^3)	514/38.5	504/39.2	505/39.2	538/36.8	19790
(11^3,5^3)	254/32	244/33.3	244/33.3	267/30.4	8133
(21^3,5^3)	1845/31.9	1761/33.4	1765/33.3	1964/29.9	58785