ICLR 2024 accepted papers on Image generation

Fast diffusion sampling · 58 papers

• Universal Guidance for Diffusion Models (148 citations)
• Effective Data Augmentation With Diffusion Models (78 citations)
• A Variational Perspective on Solving Inverse Problems with Diffusion Models (37 citations)
• DiffusionSat: A Generative Foundation Model for Satellite Imagery (31 citations)
• InstaFlow: One Step is Enough for High-Quality Diffusion-Based Text-to-Image Generation (20 citations)
• Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusion (16 citations)
• Improved Techniques for Training Consistency Models (13 citations)
• Solving Inverse Problems with Latent Diffusion Models via Hard Data Consistency (13 citations)
• Text-to-3D with Classifier Score Distillation (8 citations)
• Directly Fine-Tuning Diffusion Models on Differentiable Rewards (7 citations)
• Alleviating Exposure Bias in Diffusion Models through Sampling with Shifted Time Steps (6 citations)
• Decomposed Diffusion Sampler for Accelerating Large-Scale Inverse Problems (6 citations)

and 46 more in the interactive map.

Autoregressive and latent generation · 56 papers

• Self-Consuming Generative Models Go MAD (85 citations)
• NaturalSpeech 2: Latent Diffusion Models are Natural and Zero-Shot Speech and Singing Synthesizers (37 citations)
• Brain decoding: toward real-time reconstruction of visual perception (27 citations)
• Scalable Diffusion for Materials Generation (25 citations)
• Mixed-Type Tabular Data Synthesis with Score-based Diffusion in Latent Space (23 citations)
• MOFDiff: Coarse-grained Diffusion for Metal-Organic Framework Design (16 citations)
• Space Group Constrained Crystal Generation (15 citations)
• ScaleCrafter: Tuning-free Higher-Resolution Visual Generation with Diffusion Models (14 citations)
• Separate and Diffuse: Using a Pretrained Diffusion Model for Better Source Separation (12 citations)
• Vocos: Closing the gap between time-domain and Fourier-based neural vocoders for high-quality audio synthesis (11 citations)
• Matryoshka Diffusion Models (10 citations)
• An Efficient Membership Inference Attack for the Diffusion Model by Proximal Initialization (9 citations)

and 44 more in the interactive map.

Text-to-image diffusion · 39 papers

• SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis (322 citations)
• AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuning (85 citations)
• Training Diffusion Models with Reinforcement Learning (21 citations)
• DreamTime: An Improved Optimization Strategy for Diffusion-Guided 3D Generation (20 citations)
• DisenBooth: Identity-Preserving Disentangled Tuning for Subject-Driven Text-to-Image Generation (11 citations)
• Navigating Text-To-Image Customization: From LyCORIS Fine-Tuning to Model Evaluation (10 citations)
• AnyText: Multilingual Visual Text Generation and Editing (8 citations)
• HIFA: High-fidelity Text-to-3D Generation with Advanced Diffusion Guidance (5 citations)
• Progressive3D: Progressively Local Editing for Text-to-3D Content Creation with Complex Semantic Prompts (4 citations)
• Ring-A-Bell! How Reliable are Concept Removal Methods For Diffusion Models? (4 citations)
• The Hidden Language of Diffusion Models (4 citations)
• Würstchen: An Efficient Architecture for Large-Scale Text-to-Image Diffusion Models (4 citations)

and 27 more in the interactive map.

Image restoration · 12 papers

• Towards image compression with perfect realism at ultra-low bitrates (8 citations)
• Frequency-Aware Transformer for Learned Image Compression (7 citations)
• From Posterior Sampling to Meaningful Diversity in Image Restoration (4 citations)
• A Plug-and-Play Image Registration Network (2 citations)
• Harnessing Joint Rain-/Detail-aware Representations to Eliminate Intricate Rains (2 citations)
• Dual Associated Encoder for Face Restoration (1 citation)
• A Restoration Network as an Implicit Prior
• Idempotence and Perceptual Image Compression
• NoiseDiffusion: Correcting Noise for Image Interpolation with Diffusion Models beyond Spherical Linear Interpolation
• PAC-FNO: Parallel-Structured All-Component Fourier Neural Operators for Recognizing Low-Quality Images
• DreamClean: Restoring Clean Image Using Deep Diffusion Prior
• sRGB Real Noise Modeling via Noise-Aware Sampling with Normalizing Flows

Editing and style transfer · 12 papers

• DragonDiffusion: Enabling Drag-style Manipulation on Diffusion Models (33 citations)
• Advancing Pose-Guided Image Synthesis with Progressive Conditional Diffusion Models (17 citations)
• Image Translation as Diffusion Visual Programmers (10 citations)
• PnP Inversion: Boosting Diffusion-based Editing with 3 Lines of Code (9 citations)
• ED-NeRF: Efficient Text-Guided Editing of 3D Scene With Latent Space NeRF (8 citations)
• Unpaired Image-to-Image Translation via Neural Schrödinger Bridge (8 citations)
• The Blessing of Randomness: SDE Beats ODE in General Diffusion-based Image Editing (4 citations)
• Object-Aware Inversion and Reassembly for Image Editing (3 citations)
• Noise Map Guidance: Inversion with Spatial Context for Real Image Editing (1 citation)
• Generative Human Motion Stylization in Latent Space
• Motion Guidance: Diffusion-Based Image Editing with Differentiable Motion Estimators
• On the Analysis of GAN-based Image-to-Image Translation with Gaussian Noise Injection

Super-resolution · 4 papers

• Recursive Generalization Transformer for Image Super-Resolution (24 citations)
• Solving Diffusion ODEs with Optimal Boundary Conditions for Better Image Super-Resolution (3 citations)
• Diffusion Model for Dense Matching (2 citations)
• SEAL: A Framework for Systematic Evaluation of Real-World Super-Resolution (1 citation)

GANs · 4 papers

• Image Inpainting via Iteratively Decoupled Probabilistic Modeling (5 citations)
• SAN: Inducing Metrizability of GAN with Discriminative Normalized Linear Layer (5 citations)
• Revisit and Outstrip Entity Alignment: A Perspective of Generative Models (3 citations)
• Analyzing and Improving Optimal-Transport-based Adversarial Networks