AAAI 2025 accepted papers on Image generation
291 of the 3,486 papers accepted at AAAI 2025 (8.3%) are on Image generation, compared with 7.6% at AAAI 2024. They fall into 7 topics.
See AAAI 2025 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
Image restoration · 71 papers
- Exploiting Diffusion Prior for Real-World Image Dehazing with Unpaired Training (40 citations)
- Guided Real Image Dehazing Using YCbCr Color Space (38 citations)
- Wavelet-Assisted Multi-Frequency Attention Network for Pansharpening (33 citations)
- AGLLDiff: Guiding Diffusion Models Towards Unsupervised Training-free Real-world Low-light Image Enhancement (24 citations)
- Low-Light Image Enhancement via Generative Perceptual Priors (23 citations)
- Prior-guided Hierarchical Harmonization Network for Efficient Image Dehazing (20 citations)
- Adaptive Dual-domain Learning for Underwater Image Enhancement (18 citations)
- Dehaze-RetinexGAN: Real-World Image Dehazing via Retinex-based Generative Adversarial Network (17 citations)
- Debiased All-in-one Image Restoration with Task Uncertainty Regularization (16 citations)
- DPLUT: Unsupervised Low-light Image Enhancement with Lookup Tables and Diffusion Priors (15 citations)
- Beyond Spatial Domain: Cross-domain Promoted Fourier Convolution Helps Single Image Dehazing (12 citations)
- Multi-axis Prompt and Multi-dimension Fusion Network for All-in-one Weather-degraded Image Restoration (12 citations)
and 59 more in the interactive map.
Autoregressive and latent generation · 58 papers
- U-KAN Makes Strong Backbone for Medical Image Segmentation and Generation (308 citations)
- DiffScene: Diffusion-Based Safety-Critical Scenario Generation for Autonomous Vehicles (42 citations)
- Graphic Design with Large Multimodal Model (10 citations)
- TechSinger: Technique Controllable Multilingual Singing Voice Synthesis via Flow Matching (10 citations)
- AoP-SAM: Automation of Prompts for Efficient Segmentation (8 citations)
- Explore In-Context Segmentation via Latent Diffusion Models (8 citations)
- M2OST: Many-to-one Regression for Predicting Spatial Transcriptomics from Digital Pathology Images (8 citations)
- PVTree: Realistic and Controllable Palm Vein Generation for Recognition Tasks (8 citations)
- ODDN: Addressing Unpaired Data Challenges in Open-World Deepfake Detection on Online Social Networks (7 citations)
- DZAD: Diffusion-based Zero-shot Anomaly Detection (6 citations)
- GRPose: Learning Graph Relations for Human Image Generation with Pose Priors (6 citations)
- JEN-1 Composer: A Unified Framework for High-Fidelity Multi-Track Music Generation (6 citations)
and 46 more in the interactive map.
Editing and style transfer · 54 papers
- OOTDiffusion: Outfitting Fusion Based Latent Diffusion for Controllable Virtual Try-On (84 citations)
- IMAGDressing-v1: Customizable Virtual Dressing (78 citations)
- EchoMimic: Lifelike Audio-Driven Portrait Animations through Editable Landmark Conditions (54 citations)
- DiT4Edit: Diffusion Transformer for Image Editing (36 citations)
- Stable-Hair: Real-World Hair Transfer via Diffusion Model (13 citations)
- StableVC: Style Controllable Zero-Shot Voice Conversion with Conditional Flow Matching (12 citations)
- FaceMe: Robust Blind Face Restoration with Personal Identification (10 citations)
- SlerpFace: Face Template Protection via Spherical Linear Interpolation (10 citations)
- Color Transfer with Modulated Flows (8 citations)
- Predicting the Original Appearance of Damaged Historical Documents (8 citations)
- GIM: A Million-scale Benchmark for Generative Image Manipulation Detection and Localization (7 citations)
- HieraFashDiff: Hierarchical Fashion Design with Multi-stage Diffusion Models (7 citations)
and 42 more in the interactive map.
Text-to-image diffusion · 49 papers
- LLM4GEN: Leveraging Semantic Representation of LLMs for Text-to-Image Generation (19 citations)
- Unlearning Concepts in Diffusion Model via Concept Domain Correction and Concept Preserving Gradient (14 citations)
- MultiBooth: Towards Generating All Your Concepts in an Image from Text (11 citations)
- Perception-Guided Jailbreak Against Text-to-Image Models (11 citations)
- Resolving Multi-Condition Confusion for Finetuning-Free Personalized Image Generation (11 citations)
- DiffuseHigh: Training-Free Progressive High-Resolution Image Synthesis Through Structure Guidance (10 citations)
- CoRe: Context-Regularized Text Embedding Learning for Text-to-Image Personalization (9 citations)
- ChangeDiff: A Multi-Temporal Change Detection Data Generator with Flexible Text Prompts via Diffusion Model (8 citations)
- Revisiting Tampered Scene Text Detection in the Era of Generative AI (8 citations)
- GlyphDraw2: Automatic Generation of Complex Glyph Posters with Diffusion Models and Large Language Models (7 citations)
- ResAdapter: Domain Consistent Resolution Adapter for Diffusion Models (7 citations)
- Local Conditional Controlling for Text-to-Image Diffusion Models (6 citations)
and 37 more in the interactive map.
Fast diffusion sampling · 40 papers
- DepthFM: Fast Generative Monocular Depth Estimation with Flow Matching (37 citations)
- Boosting Consistency in Story Visualization with Rich-Contextual Conditional Diffusion Models (34 citations)
- Flash Diffusion: Accelerating Any Conditional Diffusion Model for Few Steps Image Generation (15 citations)
- Diffusion-based Synthetic Data Generation for Visible-Infrared Person Re-Identification (14 citations)
- Training-Free and Hardware-Friendly Acceleration for Diffusion Models via Similarity-based Token Pruning (10 citations)
- CasFT: Future Trend Modeling for Information Popularity Prediction with Dynamic Cues-Driven Diffusion Models (8 citations)
- Adaptive Guidance: Training-free Acceleration of Conditional Diffusion Models (6 citations)
- InpDiffusion: Image Inpainting Localization via Conditional Diffusion Models (6 citations)
- ResMaster: Mastering High-Resolution Image Generation via Structural and Fine-Grained Guidance (6 citations)
- DHMoE: Diffusion Generated Hierarchical Multi-Granular Expertise for Stock Prediction (5 citations)
- AdaDiff: Adaptive Step Selection for Fast Diffusion Models (4 citations)
- Enhancing Sequential Recommendation with Global Diffusion (4 citations)
and 28 more in the interactive map.
Super-resolution · 14 papers
- GaussianSR: High Fidelity 2D Gaussian Splatting for Arbitrary-Scale Image Super-Resolution (31 citations)
- Effective Diffusion Transformer Architecture for Image Super-Resolution (21 citations)
- Diffusion Prior Interpolation for Flexibility Real-World Face Super-Resolution (10 citations)
- EigenSR: Eigenimage-Bridged Pre-Trained RGB Learners for Single Hyperspectral Image Super-Resolution (7 citations)
- BUFF: Bayesian Uncertainty Guided Diffusion Probabilistic Model for Single Image Super-Resolution (6 citations)
- GlyphSR: A Simple Glyph-Aware Framework for Scene Text Image Super-Resolution (4 citations)
- RAP-SR: RestorAtion Prior Enhancement in Diffusion Models for Realistic Image Super-Resolution (4 citations)
- Unsupervised Degradation Representation Aware Transform for Real-World Blind Image Super-Resolution (2 citations)
- Unsupervised Diffusion-Based Degradation Modeling for Real-World Super-Resolution (2 citations)
- Block-Based Multi-Scale Image Rescaling (1 citation)
- C2PD: Continuity-Constrained Pixelwise Deformation for Guided Depth Super-Resolution (1 citation)
- Fast Omni-Directional Image Super-Resolution: Adapting the Implicit Image Function with Pixel and Semantic-Wise Spherical Geometric Priors (1 citation)
and 2 more in the interactive map.
GANs · 5 papers
- DIDiffGes: Decoupled Semi-Implicit Diffusion Models for Real-time Gesture Generation from Speech (3 citations)
- MCGAN: Enhancing GAN Training with Regression-Based Generator Loss (2 citations)
- RealisHuman: A Two-Stage Approach for Refining Malformed Human Parts in Generated Images (2 citations)
- Diverse Rare Sample Generation with Pretrained GANs
- InstantPainting: Expanding GANs for Efficient Text-Conditioned Image Generation Platform