10 datasets found

The dataset used in the paper for semantic control and manipulation of images using pre-trained diffusion models.

BioFors is a large biomedical image forensics dataset, comprising 47,805 images extracted from 1,031 open-source research papers.

The LRS2 dataset consists of 48,164 video clips from outdoor shows on BBC television. Each video is accompanied by an audio corresponding to a sentence with up to 100 characters.

Counterfactual explanations and adversarial attacks have a related goal: flipping output labels with minimal perturbations regardless of their characteristics.

The dataset used in the paper is a large dataset of images, including FFHQ, AFHQ-Cat, and LSUN-Church.

The COCO-Glide dataset is a dataset consisting of 512 manipulated images of 256 × 256 px, created using a single method (GLIDE).

i-CLEVR is a synthetic dataset generated using the CLEVR engine. Each scene contains five image-instruction pairs. Starting from a background image, new objects are added...

CoDraw is a clip-art-like dataset collected using Amazon Mechanical Turk. The images consist of combinations of 58 unique objects (such as a boy, girl, and tree) on a static...

Text-guided image manipulation tasks have recently gained attention in the vision-and-language community. The GeNeVA task is a multi-turn text-conditioned image generation...

The dataset used in the paper is a text-to-image diffusion model, specifically Stable Diffusion. The authors used this model to generate images from text prompts and evaluated...