246 datasets found

This paper proposes a framework to address the issue of requiring precise masks for existing FSS tasks, which address weakly-supervised few-shot segmentation tasks with only...

Crack detection on UAS-based Imagery using Transfer Learning

The NYU-Depth V2 dataset contains pairs of RGB and depth images collected from Microsoft Kinect in 464 indoor scenes.

A dataset for unsupervised performance evaluation of image segmentation.

The HRSOD dataset is used for evaluation, which consists of 287 samples.

The COIFT dataset is used for evaluation, which consists of 280 samples.

The DIS-VD dataset is used for evaluation, which consists of 470 samples.

The DIS-5K dataset is used for evaluation, which consists of 3,000 samples for training, 470 for validation, and 2,000 for testing (partitioned into four subsets DIS-TE1∼DIS-TE4).

UAVid is a dataset for semantic segmentation of UAV images, containing 420 images with 8 classes.

The dataset used in the paper is the Pascal VOC 2012 dataset, which is a benchmark for instance segmentation. The dataset consists of 1464 images with 20 class categories and...

The dataset used in the paper is COCO-Panoptic and ADE20K, which are widely used in the field of computer vision.

COCO Stuff dataset is an extension of the COCO dataset, 164,000 images covering 171 classes are annotated with segmentation masks.

The ISLES 2018 challenge dataset includes CT scan, CTP source data, and derived perfusion maps for ischemic stroke lesion segmentation.

A dataset of high quality is one of the key factors to train a neural network. Unfortunately, there are few open-source abdominal ultrasound image datasets.

A dataset of high quality is one of the key factors to train a neural network. Unfortunately, there are few open-source abdominal ultrasound image datasets.

The LLRGBD-synthetic dataset is a large-scale synthetic dataset with 13 semantic classes.

Understanding dark scenes based on multi-modal image data is challenging, as both the visible and auxiliary modalities provide limited semantic information for the task.

The Microsoft Common Objects in Context (COCO) dataset is a benchmark for object detection, segmentation, and image classification.

CrackSeg9k is a collection of various image datasets showing cracked and uncracked surfaces of multiple building materials.

A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics.