CVPR 2024 accepted papers on Image generation

450 of the 2,716 papers accepted at CVPR 2024 (16.6%) are on Image generation, compared with 11.6% at CVPR 2023. They fall into 7 topics.

See CVPR 2024 in the interactive map

Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.

Topics and papers

Text-to-image diffusion · 127 papers

Rethinking FID: Towards a Better Evaluation Metric for Image Generation (200 citations)
Style Aligned Image Generation via Shared Attention (131 citations)
InstantBooth: Personalized Text-to-Image Generation without Test-Time Finetuning (130 citations)
PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embedding (126 citations)
HyperDreamBooth: HyperNetworks for Fast Personalization of Text-to-Image Models (117 citations)
LEDITS++: Limitless Image Editing using Text-to-Image Models (84 citations)
InstanceDiffusion: Instance-level Control for Image Generation (82 citations)
Grounded Text-to-Image Synthesis with Attention Refocusing (74 citations)
MACE: Mass Concept Erasure in Diffusion Models (73 citations)
DEADiff: An Efficient Stylization Diffusion Model with Disentangled Representations (72 citations)
Towards Understanding Cross and Self-Attention in Stable Diffusion for Text-Guided Image Editing (72 citations)
MIGC: Multi-Instance Generation Controller for Text-to-Image Synthesis (68 citations)

and 115 more in the interactive map.

Image restoration · 81 papers

Text-IF: Leveraging Semantic Text Guidance for Degradation-Aware and Interactive Image Fusion (187 citations)
Wavelet-based Fourier Information Interaction with Frequency Diffusion Adjustment for Underwater Image Restoration (187 citations)
Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild (139 citations)
Residual Denoising Diffusion Models (138 citations)
Zero-Reference Low-Light Enhancement via Physical Quadruple Priors (135 citations)
Bidirectional Multi-Scale Implicit Neural Representations for Image Deraining (107 citations)
ZERO-IG: Zero-Shot Illumination-Guided Joint Denoising and Adaptive Enhancement for Low-Light Images (94 citations)
Fourier Priors-Guided Diffusion for Zero-Shot Joint Low-Light Enhancement and Deblurring (83 citations)
Light the Night: A Multi-Condition Diffusion Framework for Unpaired Low-Light Enhancement in Autonomous Driving (79 citations)
Efficient Multi-scale Network with Learnable Discrete Wavelet Transform for Blind Motion Deblurring (78 citations)
Selective Hourglass Mapping for Universal Image Restoration Based on Diffusion Model (74 citations)
Multimodal Prompt Perceiver: Empower Adaptiveness Generalizability and Fidelity for All-in-One Image Restoration (68 citations)

and 69 more in the interactive map.

Editing and style transfer · 76 papers

Style Injection in Diffusion: A Training-free Approach for Adapting Large-scale Diffusion Models for Style Transfer (202 citations)
DragDiffusion: Harnessing Diffusion Models for Interactive Point-based Image Editing (145 citations)
An Edit Friendly DDPM Noise Space: Inversion and Manipulations (117 citations)
Emu Edit: Precise Image Editing via Recognition and Generation Tasks (81 citations)
StegoGAN: Leveraging Steganography for Non-Bijective Image-to-Image Translation (59 citations)
Z*: Zero-shot Style Transfer via Attention Reweighting (52 citations)
HIVE: Harnessing Human Feedback for Instructional Visual Editing (50 citations)
DiffEditor: Boosting Accuracy and Flexibility on Diffusion-based Image Editing (44 citations)
CAT-DM: Controllable Accelerated Virtual Try-on with Diffusion Model (43 citations)
Coarse-to-Fine Latent Diffusion for Pose-Guided Person Image Synthesis (40 citations)
Texture-Preserving Diffusion Models for High-Fidelity Virtual Try-On (35 citations)
ZONE: Zero-Shot Instruction-Guided Local Editing (35 citations)

and 64 more in the interactive map.

Autoregressive and latent generation · 74 papers

Efficient Deformable ConvNets: Rethinking Dynamic and Sparse Operator for Vision Applications (257 citations)
Forgery-aware Adaptive Transformer for Generalizable Synthetic Image Detection (112 citations)
EditGuard: Versatile Image Watermarking for Tamper Localization and Copyright Protection (91 citations)
OmniGlue: Generalizable Feature Matching with Foundation Model Guidance (86 citations)
Gaussian Shading: Provable Performance-Lossless Image Watermarking for Diffusion Models (85 citations)
SingularTrajectory: Universal Trajectory Predictor Using Diffusion Model (67 citations)
AEROBLADE: Training-Free Detection of Latent Diffusion Images Using Autoencoder Reconstruction Error (53 citations)
DiffSHEG: A Diffusion-Based Approach for Real-Time Speech-driven Holistic 3D Expression and Gesture Generation (49 citations)
Boosting Image Quality Assessment through Efficient Transformer Adaptation with Local Feature Enhancement (48 citations)
LaRE^2: Latent Reconstruction Error Based Method for Diffusion-Generated Image Detection (48 citations)
Intrinsic Image Diffusion for Indoor Single-view Material Estimation (43 citations)
DemoFusion: Democratising High-Resolution Image Generation With No $$$ (42 citations)

and 62 more in the interactive map.

Fast diffusion sampling · 57 papers

One-step Diffusion with Distribution Matching Distillation (146 citations)
DeepCache: Accelerating Diffusion Models for Free (136 citations)
FreeU: Free Lunch in Diffusion U-Net (135 citations)
Analyzing and Improving the Training Dynamics of Diffusion Models (77 citations)
Structure Matters: Tackling the Semantic Discrepancy in Diffusion Models for Image Inpainting (76 citations)
DiffuseMix: Label-Preserving Data Augmentation with Diffusion Models (68 citations)
ECoDepth: Effective Conditioning of Diffusion Models for Monocular Depth Estimation (52 citations)
SatSynth: Augmenting Image-Mask Pairs through Diffusion Models for Aerial Semantic Segmentation (51 citations)
Learned Representation-Guided Diffusion Models for Large-Image Generation (45 citations)
DistriFusion: Distributed Parallel Inference for High-Resolution Diffusion Models (44 citations)
FlowDiffuser: Advancing Optical Flow Estimation with Diffusion Models (42 citations)
Diffusion-ES: Gradient-free Planning with Diffusion for Autonomous and Instruction-guided Driving (37 citations)

and 45 more in the interactive map.

Super-resolution · 30 papers

SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution (189 citations)
SinSR: Diffusion-Based Image Super-Resolution in a Single Step (167 citations)
Transcending the Limit of Local Window: Advanced Super-Resolution Transformer with Adaptive Token Dictionary (109 citations)
CAMixerSR: Only Details Need More "Attention" (76 citations)
CFAT: Unleashing Triangular Windows for Image Super-resolution (69 citations)
SeD: Semantic-Aware Discriminator for Image Super-Resolution (68 citations)
Rethinking Diffusion Model for Multi-Contrast MRI Super-Resolution (64 citations)
CoSeR: Bridging Image and Language for Cognitive Super-Resolution (51 citations)
Image Processing GNN: Breaking Rigidity in Super-Resolution (47 citations)
Training Generative Image Super-Resolution Models by Wavelet-Domain Losses Enables Better Control of Artifacts (45 citations)
Building Bridges across Spatial and Temporal Resolutions: Reference-Based Super-Resolution via Change Priors and Conditional Diffusion Model (38 citations)
A Dynamic Kernel Prior Model for Unsupervised Blind Image Super-Resolution (37 citations)

and 18 more in the interactive map.