Image Generation On Cifar 10

评估指标

FID

评测结果

各个模型在此基准测试上的表现结果

模型名称	FID	Paper Title	Repository
MIX-BigGAN	8.17	Lessons Learned from the Training of GANs on Artificial Datasets
ACGAN	35.47	-	-
EAGAN (G)	10.14	EAGAN: Efficient Two-stage Evolutionary Architecture Search for GANs
ProdPoly no activation functions	40.45	Deep Polynomial Neural Networks
SiDA-EDM	1.396	Adversarial Score identity Distillation: Rapidly Surpassing the Teacher in One Step
LeCAM (BigGAN + DA)	8.46	Regularizing Generative Adversarial Networks under Limited Data
Discriminator Guidance (unconditional)	1.77	Refining Generative Process with Discriminator Guidance in Score-based Diffusion Models
WGAN-GP + TT Update Rule	24.8	GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
GMem	1.22	Generative Modeling with Explicit Memory
SciRE-Solver (with EDM)	1.76	SciRE-Solver: Accelerating Diffusion Models Sampling by Score-integrand Solver with Recursive Difference	-
RESFLOW	48.29	-	-
CLR-GAN	23.3	CLR-GAN: Improving GANs Stability and Quality via Consistent Latent Representation and Reconstruction
RSGAN-GP	25.60	The relativistic discriminator: a key element missing from standard GAN
Subspace Diffusion (NSCN++)	2.17	Subspace Diffusion Generative Models
block flow	2.29	Block Flow: Learning Straight Flow on Data Blocks
StyleSAN-XL	1.36	SAN: Inducing Metrizability of GAN with Discriminative Normalized Linear Layer
RDUOT	2.95	A High-Quality Robust Diffusion Framework for Corrupted Dataset
EAGAN (G+D)	9.91	EAGAN: Efficient Two-stage Evolutionary Architecture Search for GANs
PresGAN	52.202	Prescribed Generative Adversarial Networks
PeerGAN	21.55	DuelGAN: A Duel Between Two Discriminators Stabilizes the GAN Training	-

0 of 70 row(s) selected.