Benchmark Hair Removal (GAN)

Hair Removal (GAN)

Hair Removal benchmark using hairless samples from the PH2 database and the GAN-based deep neural network by Attia et al. to simulate hair on them.

Method

Hair removal algorihm by Abbas et al. Implementation provided by the authors. Abbas, Qaisar, M. Emre Celebi, and Irene Fondón García. "Hair removal methods: a comparative study for dermoscopy images." Biomedical Signal Processing and Control 6.4 (2011): 395-404.

Results

Sample	Vifp	Uqi	Msssim	P hvs m	P hvs	Mse	Ssim	Psnr	Rmse
IMD006	0.964	0.998	0.891	29.205	28.709	64.073	0.815	30.064	8.005
IMD010	0.961	0.997	0.889	29.528	28.749	98.376	0.805	28.202	9.918
IMD017	0.957	0.994	0.918	28.996	28.385	67.983	0.820	29.807	8.245
IMD018	0.969	0.998	0.897	28.101	27.664	117.432	0.828	27.433	10.837
IMD019	0.962	0.997	0.909	27.940	27.297	90.699	0.820	28.555	9.524
IMD020	0.951	0.997	0.873	26.927	26.383	138.996	0.780	26.701	11.790
IMD030	0.961	0.996	0.919	29.119	28.429	76.885	0.807	29.272	8.768
IMD033	0.954	0.997	0.888	27.856	27.260	94.489	0.788	28.377	9.721
IMD044	0.954	0.994	0.927	29.744	29.159	78.288	0.847	29.194	8.848
IMD050	0.966	0.997	0.890	30.000	29.550	55.624	0.817	30.678	7.458
IMD061	0.932	0.993	0.842	26.186	25.351	170.740	0.731	25.807	13.067
IMD063	0.956	0.994	0.924	28.670	28.095	82.451	0.810	28.969	9.080
IMD075	0.962	0.995	0.923	28.238	27.575	82.107	0.803	28.987	9.061
Mean	0.958	0.996	0.899	28.501	27.893	93.703	0.805	28.619	9.563