Blind Face Restoration On Celeba Test

Métriques

FID

IDS

LPIPS

PSNR

SSIM

Résultats

Résultats de performance de divers modèles sur ce benchmark

						Paper Title
mGANprior	82.27	-	45.84	24.30	0.6758	Image Processing Using Multi-Code GAN Prior
HiFaceGAN	66.09	-	47.7	24.92	0.6195	HiFaceGAN: Face Renovation via Collaborative Suppression and Replenishment
DFDNet	64.65	86.21	55.4	23.15	0.629	DifFace: Blind Face Restoration with Diffused Error Contraction
CodeFormer	60.62	60	29.9	22.18	0.61	Towards Robust Blind Face Restoration with Codebook Lookup Transformer
GLEAN	60.3	67.13	46.9	23.41	0.666	DifFace: Blind Face Restoration with Diffused Error Contraction
DiffBIR	59.06	51	45.73	21.7509	0.5971	DiffBIR: Towards Blind Image Restoration with Generative Diffusion Prior
DeblurGANv2*	52.69	-	40.01	25.91	0.6952	DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and Better
PSFRGAN	52.14	68.14	50	22.74	0.63	DifFace: Blind Face Restoration with Diffused Error Contraction
PULSE	48.33	74.97	50.8	22.14	0.682	DifFace: Blind Face Restoration with Diffused Error Contraction
PSFRGAN	47.59	-	42.4	24.71	0.6557	Progressive Semantic-Aware Style Transformation for Blind Face Restoration
GFPGAN	46.99	66.76	49.5	22.18	0.631	DifFace: Blind Face Restoration with Diffused Error Contraction
VQFR	45.84	65.87	47.1	21.94	0.585	DifFace: Blind Face Restoration with Diffused Error Contraction
GFP-GAN	42.62	-	36.46	25.08	0.6777	Towards Real-World Blind Face Restoration with Generative Facial Prior
DifFace	38.43	62.39	43.5	24.08	0.703	DifFace: Blind Face Restoration with Diffused Error Contraction
PMRF	37.46	-	-	26.37	0.7073	Posterior-Mean Rectified Flow: Towards Minimum MSE Photo-Realistic Image Restoration
BDPM	12.93	-	29.3	24.58	0.742	Binary Diffusion Probabilistic Model

0 of 16 row(s) selected.

Blind Face Restoration On Celeba Test | SOTA | HyperAI