Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

SotA/Medical/Image Generation

Image Generation

567 benchmarks6689 papers

Image Generation (synthesis) is the task of generating new images from an existing dataset.

Unconditional generation refers to generating samples unconditionally from the dataset, i.e. $p(y)$
Conditional image generation (subtask) refers to generating samples conditionally from the dataset, based on a label, i.e. $p(y|x)$ .

In this section, you can find state-of-the-art leaderboards for unconditional generation. For conditional generation, and other types of image generations, refer to the subtasks.

<span style="color:grey; opacity: 0.6">( Image credit: StyleGAN )</span>

Benchmarks

Image Generation on ImageNet 256x256

FID Inception score NFE

Image Generation on GTAV-to-Cityscapes Labels

mIoU fwIOU Per-pixel Accuracy

Image Generation on CIFAR-10

FID Inception score NFE IS Intra-FID

Image Generation on COCO (Common Objects in Context)

FID Inception score FID-1 FID-2 FID-4 FID-8 SOA-C Zero shot FID

Image Generation on ImageNet 512x512

FID NFE Inception score

Image Generation on FFHQ 256 x 256

FID Precision Recall FD Coverage Density KID bits/dimension

Image Generation on ImageNet 128x128

FID Inception score IS Precision Recall

Image Generation on ImageNet 64x64

FID Bits per dim NFE Inception Score KID Inception score

Image Generation on CelebA 64x64

FID bits/dimension Precision Recall FID_CLIP Model Size (MB)

Image Generation on SYNTHIA-to-Cityscapes

MIoU (13 classes)MIoU (16 classes)mIoU (13 classes)

Image Generation on STL-10

FID Inception score Model Size (MB)Recall NFE

Image Generation on ImageNet 32x32

bpd FID Inception score

Image Generation on LSUN Churches 256 x 256

FID NFE Recall clean-FID Clean-FID (trainfull)

Image Generation on GenEval

Overall Single Obj.Two Obj.Color Attri.Colors Counting Position

Image Generation on LSUN Bedroom 256 x 256

FID FD Precision Recall FID-10k-training-steps NFE

Image Generation on FFHQ 1024 x 1024

FID bits/dimension LPIPS RMSE PSNR SSIM

Image Generation on PIE-Bench

Background PSNR Background LPIPS Structure Distance CLIPSIM

Image Generation on CelebA-HQ 256x256

FID Recall bits/dimension NFE Precision clean-FID

Image Generation on CIFAR-100

FID Inception Score Model Size (MB)Intra-FID

Image Generation on CelebA-HQ

FID P-IDS U-IDS LPIPS FLD

Image Generation on iHarmony4

Image Generation on ADE20K Labels-to-Photos

FID mIoU LPIPS Accuracy

Image Generation on COCO-Stuff Labels-to-Photos

FID mIoU Accuracy LPIPS

Image Generation on CUB

Inception score FID

Image Generation on Cityscapes Labels-to-Photo

FID mIoU Class IOU LPIPS Accuracy Per-class Accuracy Per-pixel Accuracy

Image Generation on Places2

FID P-IDS U-IDS LPIPS

Image Generation on WISE

Overall Cultural Time Space Biology Physics Chemistry

Image Generation on FFHQ-U

Image Generation on MIMIC-CXR

Image Generation on COCO 2017 val

Image Generation on Binarized MNIST

nats bits/dimension

Image Generation on CelebA 256x256

bpd FID bpd (8-bits)

Image Generation on Deep-Fashion

SSIM IS FID PCKh Retrieval Top10 Recall LPIPS DS

Image Generation on Multi-Modal-CelebA-HQ

FID LPIPS Acc Real

Image Generation on SEN12MS-CR

MAE SAM PSNR SSIM

Image Generation on CelebA-HQ 1024x1024

FID bits/dimension NFE

Image Generation on Fundus Fluorescein Angiogram Photographs & Colour Fundus Images of Diabetic Patients

FID Kernel Inception Distance

Image Generation on MNIST

FID bits/dimension Precision Recall PSNR SSIM

Image Generation on DrawBench

Aesthetics (Laion Aesthtetics Predictor)Human Preference Alignement (HPSv2)Text Alignement (SentenceBERT)

Image Generation on ADE20K-Outdoor Labels-to-Photos

FID mIoU Accuracy

Image Generation on AFHQV2

Image Generation on CelebA-HQ 128x128

FID Inception score IS

Image Generation on Fashion-MNIST

FID Precision Recall

Image Generation on HAdobe5k(1024$\times$1024)

MSE PSNR SSIM fMSE

Image Generation on LSUN Cat 256 x 256

FID Clean-FID (trainfull)

Image Generation on Oxford 102 Flowers

FID Inception score

Image Generation on Places2 val

Image Generation on SEN12MS-CR-TS

RMSE PSNR SSIM SAM

Image Generation on TextAtlasEval

StyledTextSynth OCR (F1 Score)StyledTextSynth OCR (Accuracy)StyledTextSynth OCR (Cer)StyledTextSynth FID StyledTextSynth Clip Score TextScenesHQ OCR (F1 Score)TextScenesHQ OCR (Accuracy)TextScenesHQ Clip Score TextScenesHQ OCR (Cer)TextScenesHQ FID TextVsionBlend OCR (F1 Score)TextVisionBlend OCR (Accuracy)TextVisionBlend OCR (Cer)TextVisionBlend Clip Score TextVisionBlend FID

Image Generation on cvusa

SSIM KL PSNR SD

Image Generation on AFHQ Cat

FID clean-FID MAE Signature MAE log-signature RMSE Signature RMSE log-signature clean-KID

Image Generation on ArtBench-10 (32x32)

Image Generation on CLEVR

FID-5k-training-steps

Image Generation on Cityscapes

FID-10k-training-steps

Image Generation on Cityscapes-to-Foggy Cityscapes

Image Generation on Dayton (256×256) - aerial-to-ground

SSIM KL PSNR SD

Image Generation on FFHQ

FID FID-10k-training-steps Clean-FID (70k) SSIM LPIPS PSNR

Image Generation on FFHQ 512 x 512

FID P-IDS U-IDS

Image Generation on Syn2Real-C

Image Generation on Visual Genome 64x64

FID Inception Score

Image Generation on AFHQ

Image Generation on AFHQ Dog

FID MAE Signature MAE log-signature RMSE Signature RMSE log-signature clean-FID clean-KID

Image Generation on COCO-MIG

instance success rate mIoU

Image Generation on COCO-Stuff 128x128

FID Inception Score SceneFID

Image Generation on COCO-Stuff 256x256

FID Inception Score LPIPS

Image Generation on COCO-Stuff 64x64

FID Inception Score

Image Generation on Cityscapes Photo-to-Labels

Class IOU Per-class Accuracy Per-pixel Accuracy

Image Generation on Conceptual Captions

Image Generation on Dayton (64×64) - aerial-to-ground

SSIM KL LPIPS PSNR SD

Image Generation on ImageNet

Image Generation on LLVIP

Image Generation on Places50

Image Generation on Visual Genome 128x128

FID Inception Score SceneFID

Image Generation on AFHQ Wild

FID MAE Signature MAE log-signature RMSE Signature RMSE log-signature clean-KID clean-FID

Image Generation on ARKitScenes

Image Generation on BCI

Average PSNR SSIM

Image Generation on CUB 128 x 128

FID Inception score

Image Generation on CelebA

FID PSNR SSIM bpd (8-bits)LPIPS

Image Generation on CelebA 128x128

FID Inception score MS-SSIM

Image Generation on CelebAMask-HQ

FID IS LPIPS mIoU S-FID

Image Generation on Dayton (256×256) - ground-to-aerial

Image Generation on Dayton (64x64) - ground-to-aerial

Image Generation on E.T. the Exceptional Trajectories

FD_ClaTr ClaTr-Score Classifier-F1

Image Generation on Edge-to-Handbags

Diversity Quality

Image Generation on Edge-to-Shoes

Diversity Quality

Image Generation on Ego2Top

Image Generation on FLIR

Image Generation on LSUN Horse 256 x 256

FID Clean-FID (trainfull)

Image Generation on LayoutBench-COCO - Combination

Image Generation on LayoutBench-COCO - Number

Image Generation on LayoutBench-COCO - Position

Image Generation on LayoutBench-COCO - Size

Image Generation on Pokemon 256x256

Image Generation on Replica

Image Generation on Stanford Cars

FID Inception score

Image Generation on Stanford Dogs

FID Inception score

Image Generation on VLN-CE

Image Generation on Visual Genome 256x256

FID Inception Score LPIPS

Image Generation on VizDoom

Image Generation on cat2dog

DFID FID Kernel Inception Distance

Image Generation on selfie2anime

Image Generation on ADE-Indoor

Image Generation on CAT 256x256

Image Generation on CIFAR-10 (10% data)

Image Generation on CIFAR-10 (20% data)

Image Generation on COCO

Image Generation on Cats-and-Dogs

Image Generation on CelebA-HQ 64x64

Image Generation on EPFL NIR-VIS

Image Generation on FFHQ 128 x 128

Image Generation on Freiburg Forest Dataset

Image Generation on LHQC

Image Generation on LRS2

FID LSE-D LPIPS (S1)LPIPS (S2)LPIPS (S3)LPIPS (S4)LPIPS (S5)SIFID (S1)SIFID (S2)SIFID (S3)SIFID (S4)SIFID (S5)LSE-C

Image Generation on LSUN Bedroom 64 x 64

Image Generation on LayoutBench

Image Generation on Market-1501

IS PCKh SSIM mask-IS mask-SSIM DS

Image Generation on MetFaces

MAE Signature MAE log-signature RMSE Signature RMSE log-signature

Image Generation on MetFaces-U

Image Generation on ObjectsRoom

Image Generation on Pokemon 1024x1024

Image Generation on ShapeStacks

Image Generation on Stacked MNIST

FID Inception score

Image Generation on TiO_2 nanoparticle

Image Generation on VoxCeleb2 - 1-shot learning

CSIM LPIPS Normalized Pose Error SSIM inference time (ms)FID

Image Generation on horse2zebra

Frechet Inception Distance Kernel Inception Distance Number of params

Image Generation on AFHQ-v2 64x64

Image Generation on Aerial-to-Map

Class IOU Per-class Accuracy Per-pixel Accuracy

Image Generation on BDD100K-Subsets

Image Generation on COCO-Animals

Image Generation on DPG

Image Generation on FFHQ 64x64

Image Generation on Home Action Genome

Image Generation on LAION COCO

Image Generation on LRS3

FID LSE-C LSE-D

Image Generation on LRW

FID LSE-C LSE-D LMD SSIM

Image Generation on LSUN Bedroom

FID-50k clean-FID

Image Generation on LSUN Bedroom 128 x 128

Image Generation on LSUN Car 512 x 384

Image Generation on MS-COCO

FID Inception score SOA-C

Image Generation on Object Transfiguration (sheep-to-giraffe)

classification score

Image Generation on Oxford 102 Flowers 256 x 256

Image Generation on Paris StreetView

40-50% Mask PSNR 10-20% Mask PSNR 20-30% Mask PSNR 30-40% Mask PSNR

Image Generation on RC-49

Image Generation on SYNTHIA Fall-to-Winter

mIoU fwIOU Per-pixel Accuracy

Image Generation on SYNTHIA-to-Cityscapes Labels

Image Generation on T2I-CompBench

Color Shape Texture Complex Non-Spatial Spatial

Image Generation on VoxCeleb1 - 1-shot learning

Image Generation on VoxCeleb1 - 32-shot learning

Image Generation on VoxCeleb1 - 8-shot learning

Image Generation on VoxCeleb2 - 8-shot learning

Image Generation on anime-to-selfie

Kernel Inception Distance

Image Generation on iNaturalist 2019

Image Generation on photo2vangogh

Frechet Inception Distance Kernel Inception Distance Number of params

Image Generation on selfie-to-anime

Kernel Inception Distance

Image Generation on vangogh2photo

Frechet Inception Distance Kernel Inception Distance Number of Params

Image Generation on zebra2horse

Frechet Inception Distance Kernel Inception Distance Number of params

Image Generation on 1,078 People 3D Faces Collection Data

Image Generation on 100 sleep nights of 8 caregivers

Image Generation on 2017_test set

10 way 1~2 shot

Image Generation on 25% ImageNet 128x128

Image Generation on ADE-Indoor Labels-to-Photo

Image Generation on ADE20K

Image Generation on AFHQ (Cat to Dog)

Image Generation on AFHQ (Wild to Dog)

Image Generation on ApolloScape

MAE PSNR RMSE SSIM

Image Generation on Apolloscape Inpainting

Image Generation on Apples and Oranges

Kernel Inception Distance

Image Generation on BRATS

Image Generation on CIFAR-10 LT

Image Generation on CPM-Synt-1

Image Generation on CPM-Synt-2

Image Generation on CREMA-D

EmoAcc FID LSE-C

Image Generation on CVACT

Image Generation on Cam2BEV

Image Generation on CelebA-HQ 512x512

Image Generation on Cityscapes-25K 256x512

Image Generation on Cityscapes-5K 256x512

Image Generation on Colors

Validation Accuracy

Image Generation on EMNIST-Letters

Image Generation on FFHQ 512 x 512 - 4x upscaling

Image Generation on FFHQ 64x64 - 4x upscaling

Image Generation on FFHQH

Image Generation on Flickr-8k

Image Generation on GQN

Image Generation on GeNeVA (CoDraw)

Image Generation on GeNeVA (i-CLEVR)

Image Generation on IXI

Image Generation on ImageNet 256x256 - 1 labeled data per class

FID-50k sFID IS Precision Recall

Image Generation on ImageNet 256x256 - 1% labeled data

FID-50k sFID IS Precision Recall

Image Generation on ImageNet 256x256 - 2 labeled data per class

FID-50k sFID IS Precision Recall

Image Generation on ImageNet 256x256 - 5 labeled data per class

FID-50k sFID IS Precision Recall

Image Generation on ImageNet-LT

Image Generation on Indian Celebs 256 x 256

Image Generation on KITTI Object Tracking Evaluation 2012

Image Generation on KMNIST

Image Generation on LHQ

Image Generation on LSUN

Image Generation on LSUN Car 256 x 256

Image Generation on LSUN tower 64x64

Image Generation on Landscapes 256 x 256

Image Generation on Multi-dSprites

Image Generation on NASA Perseverance

MAE Signature MAE log-signature RMSE Signature RMSE log-signature

Image Generation on Oxford 102 Flowers 128x128

Image Generation on Places365

Image Generation on SDSS Galaxies

Image Generation on Satellite-Buildings 256 x 256

Image Generation on Satellite-Landscapes 256 x 256

Image Generation on StreetView

Image Generation on Tiny ImageNet

Image Generation on VITON-HD

Image Generation on VggFace2

Image Generation on VoxCeleb2 - 32-shot learning

Image Generation on WebFace

Image Generation on Zebra and Horses

Kernel Inception Distance

Image Generation on dog2cat

Kernel Inception Distance

Image Generation on iDesigner

Image Generation on noisy-ADE20K-DS

Image Generation on noisy-ADE20K-Edge

Image Generation on noisy-ADE20K-Random

Image Generation on photo2portrait

Kernel Inception Distance

Image Generation on portrait2photo

Kernel Inception Distance

Image Generation on ‘ไอซ์ ปรีชญา’ ลืมปิดไลฟ์สดตอนอาบน้ำ คลิปถูกคนดีแชร์ออนไลน์

Image Generation on RaFD

Classification Error

Image Generation on SMPL

Image Generation on SVNH-to-MNIST

Classification Accuracy