21 datasets found

The dataset used in the experiments with the synthetic two-dimensional data and the MNIST digits.

Paired-Embedding (PE) method for effective and reliable data augmentation, Act2Act network learns from augmented data

MosaicFusion: A simple yet effective diffusion-based data augmentation approach for large vocabulary instance segmentation.

The Semantic Diffusion Model (SDM) is used for synthesizing pulmonary CT images from segmentation maps for data augmentation.

Active learning is an important technique for low-resource sequence labeling tasks. However, current active sequence labeling methods use the queried samples alone in each...

Contextual augmentation: Data augmentation by words with paradigmatic relations.

The AuGPT dataset is a task-oriented dialogue dataset with a focus on auxiliary tasks and data augmentation.

Sound event detection (SED) aims to detect sound events within an audio stream by labeling the events as well as their corresponding occurrence timestamps.

CutMix is a data augmentation method that randomly crops a patch from the source image and pastes it into the target image, forming a ground-truth label by mixing the labels of...

The dataset used in the paper is not explicitly described, but it is mentioned that the authors used it for data anonymization, data augmentation, and image perceptual quality...

A two-stage cytopathological image synthesis framework for augmenting cervical abnormality screening.

A dataset of 10k images with artificially generated soiling patterns, used for training and testing the soiling detection model.

A companion dataset to the WoodScape dataset, containing 10k images with artificially generated soiling patterns.

Face datasets are widely used in face recognition tasks, but they are often limited by their size and quality. This paper proposes a method for generating photo-realistic...

The bottle, grass, cat, and horse datasets were created for semantic segmentation tasks. The datasets contain images of 4 object types. The ND-MLS dataset was evaluated on...

The Omniglot dataset consists of 100 classes, each containing 20 images. Ten images were taken from each class for augmentation, and the rest were used as the test set. Each...

The dataset used for the experiments with the proposed approach to augment image data for semantic segmentation networks by applying image-to-image translation with both, a...

The authors propose a novel data augmentation method called Cross Domain Generative Augmentation (CDGA) to reduce the estimation error of Empirical Risk Minimization (ERM) under...

Self-supervised monocular depth estimation has gathered no-table interest since it can liberate training from dependency on depth annotations. In monocular video training case,...

MixGen: a joint data augmentation for vision-language representation learning to further improve data efficiency.