Image Generation

OSEDiff: One-step diffusion network for real-world image super-resolution, achieving comparable or better results than multi-step methods with significantly reduced computational cost and improved ima…

Score Implicit Matching (SIM) revolutionizes diffusion model distillation by creating high-quality, single-step generators from complex, multi-step models, achieving comparable performance and enablin…

Researchers achieve state-of-the-art image generation by disentangling semantic and control metadata in diffusion models and optimizing pre-training across resolutions.

Unlocking superior discriminative features from diffusion models, this research reveals key activation properties for effective feature selection, surpassing state-of-the-art methods.

Neural-RDM: A novel framework for deep, scalable vision generation using residual diffusion models, achieving state-of-the-art results on image and video benchmarks.

Neural Gaffer: Relighting any object via diffusion using a single image and an environment map to produce high-quality, realistic relit images.

This study introduces a neural cover selection framework for image steganography, optimizing latent spaces in generative models to improve message recovery and image quality.

Neural Assets enables intuitive 3D multi-object scene editing via image diffusion models by using per-object representations to control individual object poses, achieving state-of-the-art results.

New method distills slow diffusion models into fast, few-step models by matching data expectations, achieving state-of-the-art results on ImageNet.

MonkeySee reconstructs natural images from macaque brain signals with high accuracy using a novel CNN decoder, advancing neural decoding and offering insights into visual perception.

MimicTalk generates realistic, expressive talking videos in minutes using a pre-trained model adapted to individual identities.

MC-DiT: A novel training paradigm for masked diffusion models achieving state-of-the-art image generation by leveraging clean-to-clean reconstruction.

Masked Pre-training empowers a universal, fast zero-shot image denoiser!

CATOD framework improves text-to-image generation by actively learning high-quality training data to accurately depict out-of-distribution concepts.

LucidDrag: Semantic-aware dragging transforms image editing with an intention reasoner and collaborative guidance, achieving superior accuracy, image fidelity, and semantic diversity.

LiteVAE: A new autoencoder design for latent diffusion models boosts efficiency sixfold without sacrificing image quality, achieving faster training and lower memory needs via the 2D discrete wavelet …

Learning-to-Cache (L2C) dramatically accelerates diffusion transformers by intelligently caching layer computations, achieving significant speedups with minimal performance loss.

STRAINER: A new framework enabling faster, higher-quality INR fitting by leveraging transferable features across similar signals, significantly boosting INR performance.

PaDIS: Patch-based diffusion inverse solver learns efficient image priors from image patches, enabling high-resolution inverse problem solutions with reduced computational costs and data needs.

This paper proposes a novel method to model group actions within autoencoders by learning these actions in the latent space, enhancing model versatility and improving performance in various real-world…