Image Generation

Dual-Stream Interactive Transformers (DSIT) revolutionizes single image reflection separation by using a novel dual-attention mechanism that captures inter- and intra-layer correlations, significantly…

Simple and Fast Distillation (SFD) drastically accelerates diffusion model training by 1000x, achieving state-of-the-art results in few-step image generation with minimal fine-tuning.

ShowMaker: Generating high-fidelity 2D human conversational videos using fine-grained diffusion modeling and 2D key points.

SHMT: Self-supervised Hierarchical Makeup Transfer uses latent diffusion models to realistically and precisely apply diverse makeup styles to faces, even without paired training data, achieving high f…

SemFlow: A unified framework uses rectified flow to seamlessly bridge semantic segmentation and image synthesis, achieving competitive results and offering reversible image-mask transformations.

Self-Play Fine-Tuning (SPIN-Diffusion) revolutionizes diffusion model training, achieving superior text-to-image results with less data via iterative self-improvement, surpassing supervised and RLHF m…

Researchers improved 3D shape generation from 2D diffusion models by showing that existing Score Distillation Sampling is a reparameterized version of DDIM and fixing its high-variance noise issue via…

Logistic Schedule: A novel noise schedule revolutionizes image editing by improving DDIM inversion, enhancing content preservation and edit fidelity without model retraining!

DisCo: a novel framework for generalizable complex image generation using scene graph disentanglement and composition, achieving superior performance over existing methods.

VQGAN-LC massively scales VQGAN’s codebook to 100,000 entries while maintaining a 99% utilization rate, significantly boosting image generation and downstream task performance.

ROBIN: A novel watermarking method for diffusion models that actively conceals robust watermarks using adversarial optimization, enabling strong, imperceptible, and verifiable image authentication.

Revolutionizing image generation, Representation-Conditioned Generation (RCG) achieves state-of-the-art results in unconditional image synthesis by leveraging self-supervised representations to condit…

This paper presents a novel regional cross-attention module for rich-context layout-to-image generation, significantly improving image accuracy while addressing limitations of existing methods. Two n…

Researchers enhanced image generation by improving score distillation sampling via a novel Schrödinger Bridge framework, improving realism without computational overhead.

Revolutionizing image exposure assessment, Pixel-level IEA Network (P-IEANet) achieves state-of-the-art performance with a novel pixel-level approach, a new dataset (IEA40K), and a benchmark of 19 met…

WBSR: A novel framework for efficient image super-resolution that tackles data and model imbalances for superior performance and approximately a 34% reduction in computational cost.

RestoreAgent, an AI-powered image restoration agent, autonomously identifies and corrects multiple image degradations, exceeding human expert performance.

Resfusion, a novel framework, accelerates image restoration by integrating residual noise into the diffusion process, achieving superior results with fewer steps.

ReNO: Boost one-step text-to-image models by cleverly optimizing initial noise using reward signals, achieving state-of-the-art results efficiently.

Remix-DiT: Boosting diffusion model image generation quality by cleverly mixing smaller basis models into numerous specialized denoisers, improving efficiency and lowering costs!