135 datasets found

The WildDash2 dataset consists of 3,618 training, 638 validation, and 812 test images, containing a total of 25 classes.

State-of-the-art techniques in weakly-supervised semantic segmentation (WSSS) using image-level labels exhibit severe performance degradation on driving scene datasets such as...

Cityscapes dataset is a large-scale urban scene dataset containing 30,000 images of street scenes.

A benchmark Abdominal Adipose Tissue CT Image Dataset (AATTCT-IDS) for image denoising, semantic segmentation, and radiomics evaluation.

The dataset used in this paper for 3D point cloud classification and semantic segmentation tasks.

Few-shot segmentation aims at performing semantic segmentation on novel classes given a few annotated support samples.

MF: A proprietary mobile mapping platform dataset.

TUM-MLS-2016: An annotated mobile LiDAR dataset of the TUM City Campus for semantic point cloud interpretation in urban areas.

Semantic image segmentation plays a pivotal role in many vision applications including autonomous driving and medical image analysis.

The dataset used in this paper is ImageNet and CIFAR-10 for image classification, and Cityscapes for semantic segmentation.

The dataset used in the paper is ScanNetV2, a real-world dataset for indoor scenes. It contains 1205 training scenes and 312 testing scenes, with instance-level object bounding...

The dataset used in this paper is ADE20k for semantic segmentation.

The Cityscapes dataset is a large-scale urban scene dataset containing over 25,000 images.

The 2D-3D-S dataset is an indoor dataset with multiple modalities from 2D, 2.5D and 3D domains, with instance-level semantic and geometric annotations.

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 ImageNet-1K, ADE20K, and COCO 2017.

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

Learning from bounding-boxes annotations has shown great potential in weakly-supervised 3D point cloud instance segmentation. However, we observed that existing methods would...

Pyramid scene parsing network for semantic segmentation.