Conditional Image Generation On Imagenet

FID

Inception score

평가 결과

이 벤치마크에서 각 모델의 성능 결과

비교 표

모델 이름	FID	Inception score
omni-gan-on-the-secrets-of-cgans-and-beyond	8.30	190.94
simple-diffusion-end-to-end-diffusion-for	3.23	171.9
cgans-with-projection-discriminator	27.62	36.8
feature-quantization-improves-gan-training	13.77	54.36
entropy-driven-sampling-and-training-scheme	2.63	159.72
large-scale-gan-training-for-high-fidelity	5.7	124.5
large-scale-gan-training-for-high-fidelity	8.7	98.8
conditional-image-synthesis-with-auxiliary	-	28.5
high-fidelity-image-generation-with-fewer	7.7	83.1
instance-selection-for-gans	9.61	114.32
entropy-driven-sampling-and-training-scheme	2.68	169.24
self-attention-generative-adversarial	18.65	52.52
your-local-gan-designing-two-dimensional	15.94	57.22
elucidating-the-design-space-of-classifier	2.19	-
omni-gan-on-the-secrets-of-cgans-and-beyond	6.53	262.85
consistency-regularization-for-generative-1	6.66	-
instance-conditioned-gan	9.5	108.6
cgans-with-auxiliary-discriminative	8.02	108.10
is-attention-better-than-matrix-decomposition-1	14.80	58.75
diffusion-models-beat-gans-on-image-synthesis	2.97	-
rebooting-acgan-auxiliary-classifier-gans	8.206	96.299
simple-diffusion-end-to-end-diffusion-for	2.88	137.3