65 datasets found

A semi-self-supervised domain adaptation technique based on deep convolutional neural networks with a probabilistic diffusion process, requiring minimal manual data annotation.

VisDA dataset consists of 12 categories and contains around 150,000 synthetic and 50,000 real-world images in the source and target domains, respectively.

Office dataset consists of 31 categories and contains around 4,700 images across three domains, namely Amazon(A), DSLR(D) and Webcam(W).

The dataset used in the paper is the dSprites dataset, which contains 2D-shape binary images with a size of 64×64, synthetically generated with five independent factors: shape...

The Domain Adaptation with Geometric Preservation and Distribution Alignment (DAGDA) dataset is a dataset for domain adaptation with geometric preservation and distribution...

The Conditional Adversarial Domain Adaptation (CADAN) dataset is a dataset for conditional adversarial domain adaptation.

The MIMIC-CXR dataset is currently the largest radiology imaging dataset available. It comprises a total of 473,057 chest X-ray images and 206,563 associated radiology...

GeneIC uses CUB-200 and Oxford-102 datasets for training and testing.

Fundus image dataset for unsupervised domain adaptive segmentation

Fundus image dataset for unsupervised domain adaptive segmentation

VisDA-C is a dataset for visual domain adaptation, consisting of synthetic rendering of 3D models in the source domain and Microsoft COCO real images in the target domain.

Source-free domain adaptation (SFDA) aims to adapt a source model trained on a fully-labeled source domain to a related but unlabeled target domain.

This paper tackles a novel problem: how to transfer knowledge from the emerging Segment Anything Model (SAM) to learn a compact panoramic semantic segmentation model, i.e.,...

The Amazon review dataset is used for multi-source domain adaptation. It contains review texts and ratings of bought products. Products are grouped into categories. Following...

The dataset used in the paper is a multi-view clustering dataset, which contains 6 views of 30000 samples each. The dataset is used to evaluate the performance of the proposed...

The SYNTHIA dataset is a synthetic dataset for semantic segmentation, and the Cityscapes dataset is a real-world dataset for semantic segmentation.

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.

Office-Home, Office-31, and DomainNet are benchmark datasets for semi-supervised domain adaptation.

VisDA-2017 is a simulation-to-real dataset with two extremely distinct domains: Synthetic renderings of 3D models and Real collected from photo-realistic or real-image datasets.

Unsupervised Domain Adaptation (UDA) aims to leverage a label-rich source domain to solve tasks on a related unlabeled target domain.