HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Image To Image Translation On Ade20K Labels

평가 지표

FID

mIoU

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	FID	mIoU	Paper Title	Repository
USIS-Wavelet	34.5	16.95	Wavelet-based Unsupervised Label-to-Image Translation
CRN	73.3	22.4	Photographic Image Synthesis with Cascaded Refinement Networks	-
CoCosNet	26.4	-	Cross-domain Correspondence Learning for Exemplar-based Image Translation
INADE	-	-	Diverse Semantic Image Synthesis via Probability Distribution Modeling
OASIS	28.3	48.8	You Only Need Adversarial Supervision for Semantic Image Synthesis
DP-GAN	26.1	52.7	Dual Pyramid Generative Adversarial Networks for Semantic Image Synthesis
pix2pixHD	81.8	20.3	High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs
DP-SIMS (ConvNext-L)	22.7	54.3	Unlocking Pre-trained Image Backbones for Semantic Image Synthesis	-
SPADE	33.9	38.5	Semantic Image Synthesis with Spatially-Adaptive Normalization
VQGAN+Transformer	35.5	-	Taming Transformers for High-Resolution Image Synthesis
SPADE + SESAME	31.9	49	SESAME: Semantic Editing of Scenes by Adding, Manipulating or Erasing Objects
Pix2PixHD-AUG	41.5	-	Improving Augmentation and Evaluation Schemes for Semantic Image Synthesis	-
CC-FPSE-AUG	32.6	44	Improving Augmentation and Evaluation Schemes for Semantic Image Synthesis	-
CC-FPSE	31.7	43.7	Learning to Predict Layout-to-image Conditional Convolutions for Semantic Image Synthesis
SCDM	26.9	49.4	Stochastic Conditional Diffusion Models for Robust Semantic Image Synthesis
USIS	33.2	17.38	USIS: Unsupervised Semantic Image Synthesis

0 of 16 row(s) selected.