ICCV 2023 accepted papers on Image generation
258 of the 2,156 papers accepted at ICCV 2023 (12.0%) are on Image generation, compared with 11.0% at ICCV 2021. They fall into 7 topics.
See ICCV 2023 in the interactive map
Submitting to ICCV? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Image restoration · 77 papers
- Retinexformer: One-stage Retinex-based Transformer for Low-light Image Enhancement (1,086 citations)
- DiffIR: Efficient Diffusion Model for Image Restoration (479 citations)
- Diff-Retinex: Rethinking Low-light Image Enhancement with A Generative Diffusion Model (335 citations)
- Focal Network for Image Restoration (323 citations)
- Implicit Neural Representation for Cooperative Low-light Image Enhancement (296 citations)
- Parallax-Tolerant Unsupervised Deep Image Stitching (144 citations)
- Multiscale Structure Guided Diffusion for Image Deblurring (119 citations)
- ExposureDiffusion: Learning to Expose for Low-light Image Enhancement (109 citations)
- DDColor: Towards Photo-Realistic Image Colorization via Dual Decoders (107 citations)
- DDS2M: Self-Supervised Denoising Diffusion Spatio-Spectral Model for Hyperspectral Image Restoration (89 citations)
- Pixel Adaptive Deep Unfolding Transformer for Hyperspectral Image Reconstruction (86 citations)
- COOL-CHIC: Coordinate-based Low Complexity Hierarchical Image Codec (84 citations)
and 65 more in the interactive map.
Text-to-image diffusion · 63 papers
- Adding Conditional Control to Text-to-Image Diffusion Models (6,074 citations)
- MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editing (604 citations)
- ELITE: Encoding Visual Concepts into Textual Embeddings for Customized Text-to-Image Generation (391 citations)
- Erasing Concepts from Diffusion Models (373 citations)
- Human Preference Score: Better Aligning Text-to-Image Models with Human Preference (338 citations)
- SVDiff: Compact Parameter Space for Diffusion Fine-Tuning (303 citations)
- DALL-Eval: Probing the Reasoning Skills and Social Biases of Text-to-Image Generation Models (298 citations)
- Unleashing Text-to-Image Diffusion Models for Visual Perception (250 citations)
- Your Diffusion Model is Secretly a Zero-Shot Classifier (243 citations)
- Iterative Prompt Learning for Unsupervised Backlit Image Enhancement (237 citations)
- DreamBooth3D: Subject-Driven Text-to-3D Generation (236 citations)
- BoxDiff: Text-to-Image Synthesis with Training-Free Box-Constrained Diffusion (221 citations)
and 51 more in the interactive map.
Editing and style transfer · 38 papers
- StyleDiffusion: Controllable Disentangled Style Transfer via Diffusion Models (230 citations)
- Multimodal Garment Designer: Human-Centric Latent Diffusion Models for Fashion Image Editing (114 citations)
- A Latent Space of Stochastic Diffusion Models for Zero-Shot Image Editing and Guidance (97 citations)
- Prompt Tuning Inversion for Text-driven Image Editing Using Diffusion Models (82 citations)
- AesPA-Net: Aesthetic Pattern-Aware Style Transfer Networks (78 citations)
- BlendFace: Re-designing Identity Encoders for Face-Swapping (78 citations)
- Adverse Weather Removal with Codebook Priors (65 citations)
- All-to-Key Attention for Arbitrary Style Transfer (63 citations)
- HyperReenact: One-Shot Reenactment via Jointly Learning to Refine and Retarget Faces (51 citations)
- Effective Real Image Editing with Accelerated Iterative Diffusion Inversion (47 citations)
- Few Shot Font Generation Via Transferring Similarity Guided Global Style and Quantization Local Style (40 citations)
- Controllable Person Image Synthesis with Pose-Constrained Latent Diffusion (37 citations)
and 26 more in the interactive map.
Super-resolution · 28 papers
- Dual Aggregation Transformer for Image Super-Resolution (527 citations)
- SRFormer: Permuted Self-Attention for Single Image Super-Resolution (423 citations)
- EfficientViT: Lightweight Multi-Scale Attention for High-Resolution Dense Prediction (396 citations)
- Spatially-Adaptive Feature Modulation for Efficient Image Super-Resolution (323 citations)
- ESSAformer: Efficient Transformer for Hyperspectral Image Super-resolution (189 citations)
- Feature Modulation Transformer: Cross-Refinement of Global Representation via High-Frequency Prior for Image Super-Resolution (155 citations)
- HSR-Diff: Hyperspectral Image Super-Resolution via Conditional Diffusion Models (119 citations)
- Learning Non-Local Spatial-Angular Correlation for Light Field Image Super-Resolution (115 citations)
- Lightweight Image Super-Resolution with Superpixel Token Interaction (79 citations)
- Content-Aware Local GAN for Photo-Realistic Super-Resolution (74 citations)
- CuNeRF: Cube-Based Neural Radiance Field for Zero-Shot Medical Image Arbitrary-Scale Super Resolution (51 citations)
- Decomposition-Based Variational Network for Multi-Contrast MRI Super-Resolution and Reconstruction (50 citations)
and 16 more in the interactive map.
Autoregressive and latent generation · 22 papers
- Scalable Diffusion Models with Transformers (3,816 citations)
- DDFM: Denoising Diffusion Model for Multi-Modality Image Fusion (474 citations)
- DIRE for Diffusion-Generated Image Detection (468 citations)
- The Stable Signature: Rooting Watermarks in Latent Diffusion Models (337 citations)
- Masked Diffusion Transformer is a Strong Image Synthesizer (114 citations)
- Denoising Diffusion Autoencoders are Unified Self-supervised Learners (88 citations)
- Diffusion Models as Masked Autoencoders (80 citations)
- Diffusion Model as Representation Learner (77 citations)
- Removing Anomalies as Noises for Industrial Defect Localization (68 citations)
- SeeABLE: Soft Discrepancies and Bounded Contrastive Learning for Exposing Deepfakes (60 citations)
- SALAD: Part-Level Latent Diffusion for 3D Shape Generation and Manipulation (49 citations)
- LivelySpeaker: Towards Semantic-Aware Co-Speech Gesture Generation (46 citations)
and 10 more in the interactive map.
Fast diffusion sampling · 20 papers
- DiffuMask: Synthesizing Images with Pixel-level Annotations for Semantic Segmentation Using Diffusion Models (229 citations)
- Q-Diffusion: Quantizing Diffusion Models (217 citations)
- DDP: Diffusion Model for Dense Visual Prediction (202 citations)
- Efficient Diffusion Training via Min-SNR Weighting Strategy (189 citations)
- FreeDoM: Training-Free Energy-Guided Conditional Diffusion Model (184 citations)
- DOLCE: A Model-Based Probabilistic Diffusion Framework for Limited-Angle CT Reconstruction (122 citations)
- Improving Sample Quality of Diffusion Models Using Self-Attention Guidance (120 citations)
- HyperDiffusion: Generating Implicit Neural Fields with Weight-Space Diffusion (101 citations)
- Score-Based Diffusion Models as Principled Priors for Inverse Imaging (100 citations)
- LayoutDiffusion: Improving Graphic Layout Generation by Discrete Diffusion Probabilistic Models (56 citations)
- Stochastic Segmentation with Conditional Categorical Diffusion Models (49 citations)
- Diffusion in Style (46 citations)
and 8 more in the interactive map.
GANs · 10 papers
- Unsupervised Image Denoising in Real-World Scenarios via Self-Collaboration Parallel Generative Adversarial Branches (51 citations)
- MI-GAN: A Simple Baseline for Image Inpainting on Mobile Devices (42 citations)
- MosaiQ: Quantum Generative Adversarial Networks for Image Generation on NISQ Computers (36 citations)
- GlowGAN: Unsupervised Learning of HDR Images from LDR Images in the Wild (25 citations)
- LFS-GAN: Lifelong Few-Shot Image Generation (24 citations)
- DreamTeacher: Pretraining Image Backbones with Deep Generative Models (19 citations)
- What can Discriminator do? Towards Box-free Ownership Verification of Generative Adversarial Networks (15 citations)
- Dark Side Augmentation: Generating Diverse Night Examples for Metric Learning (9 citations)
- Neural Characteristic Function Learning for Conditional Image Generation (6 citations)
- Conceptual and Hierarchical Latent Space Decomposition for Face Editing (5 citations)
