SOTAVerified

Image Generation

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.

( Image credit: StyleGAN )

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 5451–5475 of 6689 papers

TitleStatusHype
EC-GAN: Low-Sample Classification using Semi-Supervised Algorithms and GANsCode1
Soft-IntroVAE: Analyzing and Improving the Introspective Variational AutoencoderCode1
Evaluating GAN-Based Image Augmentation for Threat Detection in Large-Scale Xray Security Images—0
Focal Frequency Loss for Image Reconstruction and SynthesisCode1
Correspondence Learning for Controllable Person Image Generation—0
Vid2Actor: Free-viewpoint Animatable Person Synthesis from Video in the Wild—0
GuidedStyle: Attribute Knowledge Guided Style Manipulation for Semantic Face Editing—0
High-Fidelity Neural Human Motion Transfer from Monocular VideoCode1
Self-supervised monocular depth estimation from oblique UAV videosCode0
Three Dimensional MR Image Synthesis with Progressive Generative Adversarial Networks—0
Infinite Nature: Perpetual View Generation of Natural Scenes from a Single ImageCode0
Unlabeled Data Guided Semi-supervised Histopathology Image Segmentation—0
Taming Transformers for High-Resolution Image SynthesisCode1
Combating Mode Collapse in GAN training: An Empirical Analysis using Hessian Eigenvalues—0
Self-Supervised Sketch-to-Image SynthesisCode1
Learning Self-Consistency for Deepfake DetectionCode1
Latent Space Conditioning on Generative Adversarial Networks—0
Learning Energy-Based Models by Diffusion Recovery LikelihoodCode1
Human Pose Transfer by Adaptive Hierarchical DeformationCode1
Sat2Vid: Street-view Panoramic Video Synthesis from a Single Satellite Image—0
Learning Energy-Based Models With Adversarial TrainingCode0
Full-Glow: Fully conditional Glow for more realistic image generationCode1
Enhance Convolutional Neural Networks with Noise Incentive Block—0
Improving the Fairness of Deep Generative Models without RetrainingCode1
Conditional Generation of Medical Images via Disentangled Adversarial Inference—0
Show:102550
← PrevPage 219 of 268Next →
All datasetsImageNet 256x256CIFAR-10ImageNet 64x64ImageNet 512x512FFHQ 256 x 256CelebA 64x64ImageNet 32x32LSUN Bedroom 256 x 256STL-10LSUN Churches 256 x 256ImageNet 128x128FFHQ 1024 x 1024

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Improved DDPMFID12.3—Unverified
2ADMFID11.84—Unverified
3BigGAN-deepFID8.1—Unverified
4Polarity-BigGANFID6.82—Unverified
5VQGAN+Transformer (k=mixed, p=1.0, a=0.005)FID6.59—Unverified
6MaskGITFID6.18—Unverified
7VQGAN+Transformer (k=600, p=1.0, a=0.05)FID5.2—Unverified
8CDMFID4.88—Unverified
9ADM-GFID4.59—Unverified
10RINFID4.51—Unverified
#ModelMetricClaimedVerifiedStatus
1PresGANFID52.2—Unverified
2RESFLOWFID48.29—Unverified
3Residual FlowFID46.37—Unverified
4GLF+perceptual loss (ours)FID44.6—Unverified
5ProdPoly no activation functionsFID40.45—Unverified
6ProdPoly no activation functionsFID36.77—Unverified
7ACGANFID35.47—Unverified
8DenseFlow-74-10FID34.9—Unverified
9NVAE w/ flowFID32.53—Unverified
10QSNGANFID31.97—Unverified
#ModelMetricClaimedVerifiedStatus
1GLIDE + CLSFID30.87—Unverified
2GLIDE + CLIPFID30.46—Unverified
3GLIDE + CLS-FREEFID29.22—Unverified
4GLIDE + CLIP + CLS + CLS-FREEFID29.18—Unverified
5PGMGANFID21.73—Unverified
6CLR-GANFID20.27—Unverified
7FMFID14.45—Unverified
8CT (Direct Generation, NFE=1)FID13—Unverified
9CT (Direct Generation, NFE=2)FID11.1—Unverified
10GLIDE +CLSKID7.95—Unverified