82 datasets found

The dataset used in the paper is LAION-2B, which is a large-scale image-text dataset. The authors fine-tune a pre-trained diffusion model with a subset of LAION-2B with 10k...

The MVDream dataset contains images with various objects and scenes.

The LAION5B dataset contains images with various objects and scenes.

The dataset contains 1000 pairs of images for learning and 50 pairs for evaluation.

The dataset used in this paper is DreamBench, which contains 30 different subject images including backpacks, sneakers, boots, cats, dogs, and toy, etc.

Google Open Images Dataset used to obtain licence plate images.

The dataset used in this paper is Symsol_reduced and Celeba.

The dataset used in this paper is dSprites2, Cars3D, and SmallNorb.

The dataset used in the paper is a collection of images from the SVHN dataset, along with labels. The dataset is used for image classification.

The OfficeHome dataset is a benchmark for domain generalization, containing 13,000 images from 6 domains: Clipart, Product, Real World, Art, Product, and Real World.

The MSCOCO 2014 validation dataset is used to evaluate the performance of the proposed self-perceptual objective.

The dataset used in the paper is YFCC100M, a large-scale video dataset. The dataset is used for foreground and background patch extraction and object recognition tasks.

Low-light image enhancement task using a generative diffusion model and Retinex decomposition

The unlabeled dataset contained 517 RGB-D apple tree images taken from a commercial apple orchard in Victoria, Australia.

Object recognition is arguably the most important problem at the heart of computer vision. Recently, Barbu et al. introduced a dataset called ObjectNet which includes objects in...

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.

Large-scale vision-and-language models trained on curated and web-scrapped data have led to significant improvements over task-specific models when transferred to downstream...

The Visual Genome dataset is a large-scale visual question answering dataset, containing 1.5 million images, each with 15-30 annotated entities, attributes, and relationships.

Face recognition and person re-identification using paired image-attribute data, where the attributes (i.e., soft biometrics) are only available during the training phase.

Large scale datasets [18, 17, 27, 6] boosted text conditional image generation quality. However, in some domains it could be difficult to make such datasets and usually it could...