190 datasets found

The dataset used in the paper for preference guided image generation with relative attributes.

The dataset used in the paper is the Flickr-Faces-HQ Dataset (FFHQ) and CelebA dataset.

The DrawBench dataset is a benchmark for text-to-image synthesis, consisting of images with multiple subjects and objects.

MNIST Fashion and CIFAR-10 are two well-known balanced datasets, MNIST Fashion and CIFAR-10. We first sample 70% of images as the training set for generative models.

Generative adversarial networks (GANs) are one of the most powerful generative models, but always require a large and balanced dataset to train. Traditional GANs are not...

The GEMRec-18K dataset contains 18 thousand images generated by 200 text-to-image diffusion models fine-tuned on Stable Diffusion.

The dataset used in the paper is a large-scale image generation dataset, which is used to evaluate the performance of the characteristic guidance method.

The dataset used in the paper is a large-scale image generation dataset, which is used to evaluate the performance of the characteristic guidance method.

LazyDiffusion is a novel transformer-based encoder-decoder architecture for interactive image generation and editing using a diffusion model.

Laion-Aesthetics-V2 is a large-scale dataset for training next-generation image-text models.

UniControl is a unified diffusion model for controllable visual generation in the wild, which is capable of simultaneously handling various visual conditions for the...

The AmbientGAN model adapts the original GAN configuration to handle cases with noisy or incomplete samples.

The COCO dataset is a large-scale dataset for object detection, semantic segmentation, and captioning. It contains 80 object categories and 1,000 image instances per category,...

Progress in GANs has enabled the generation of high-res-olution photorealistic images of astonishing quality. StyleGANs allow for compelling attribute modification on such...

Model inversion attacks are a type of privacy attack that reconstructs private data used to train a machine learning model, solely by accessing the model.

The dataset used in this paper is a mixture of Gaussian distributions (MoG) with 4 components, each with a 2D Gaussian distribution with means [-1, -1], [-1, 1], [1, 1], [1, -1]...

A synthetic image dataset generated with Stable Diffusion using the Wordnet taxonomy and the definitions of concepts it contains.

Visual AutoRegressive modeling (VAR), a new generation paradigm that redefines the autoregressive learning on images as coarse-to-fine “next-scale prediction” or...

The dataset used in the paper is not explicitly described, but it is mentioned that it is a large-scale captioned image dataset (LAION) used to train the Stable Diffusion model.

The COCO2017 dataset is used for the trainability experiment, which includes 47,429 images with 210,893 objects.