135 datasets found

Diff2Scene uses ScanNet, Matterport3D, ScanNet200 and Replica for open-vocabulary 3D semantic segmentation and visual grounding tasks.

The dataset used in the paper is a synthetic dataset for image-based object detection tasks, specifically instance segmentation and monocular 3D detection.

SemanticKITTI: A dataset for semantic scene understanding of lidar sequences.

A large-scale synthetic dataset for LiDAR semantic segmentation, consisting of 13 LiDAR point cloud sequences with 198,396 scans in total.

The dataset used in the paper is the COCO object detection and instance segmentation dataset, and the ADE20K semantic segmentation dataset.

The VOC2012 dataset is a multi-label image recognition dataset. It contains 11,540 train-val images and 10,991 test images.

ADE20K: A Dataset for Semantic Segmentation

Weakly-supervised semantic segmentation using plain Vision Transformer (ViT) for Weakly-supervised Semantic Segmentation (WSSS).

SemanticKITTI and SemanticPOSS are large-scale outdoor scene LiDAR datasets, which provide the complete point-wise labels for all 22 sequences in KITTI Odometry Benchmark.

NYUD2 dataset is a dataset containing RGB-D images for indoor scene understanding. It is used for depth estimation and semantic segmentation tasks.

CS-150 is a dataset consisting of 150 papers and includes 1175 images.

DSSE-200 is a complex document layout dataset including various dataset styles.

Document layout analysis (DLA) aims to split the document image into different interest regions and understand the role of each region.

The ADE20K dataset is a large-scale dataset for semantic segmentation. It contains 20,000 images with 150 semantic categories, with 20,000 images for training, 2,000 images for...

Convolutional Neural Networks (CNNs) have performed extremely well on data represented by regularly arranged grids such as images. However, directly leveraging the classic...

The CrackMap dataset is a new publicly available dataset for road crack segmentation.

The CAMELYON16 dataset consists of 399 Hematoxylin & Eosin (H&E)-stained WSIs of sentinel axillary lymph node sections. The training set contains 111 WSIs with and 159...

Real-time semantic segmentation system for high-resolution images

The Camvid and Cityscapes datasets are used for semantic segmentation tasks.

CamVid dataset is a benchmark dataset for semantic segmentation. It consists of 700 images with 11 object classes.