*.png

Glacial lake outburst floods (GLOFs) are a major threat to the local communities and important infrastructures in the high mountain regions. Early detection of glacial lakes can prevent these disastrous events. Towards this end, we collected Sentinel 2 true color scenes of High-Mountain Asia (HMA) region using glacial lakes inventory of this region. It covers an area of 2080.12 km2 with nearly 30,121 glacial lakes. After data collection, we retained 1200 cloud free true color images and manually generated their ground truth masks.

The dataset contains maps with the objective analysis (OA) fields of available water contents.  OA is based on ground-base and remote sensing (ASCAT) observations in the upper 10 and 20 cm soil layers for the period from September 11 to November 16, 2022. 

It presents bigest Assamese Isolated handwritten numeral dataset that is consist of   35106 numerals samples in grayscale format of uniform size 28x28. The contributors arestudents and Academic professionals.

This dataset consists of 1878 labeled images of flowers from blackberry trees from the specie Rubus L. subgenus Rubus Watson. These are white flowers with five petals that blossom in the spring through summer. The images were collected using an Intel RealSense D435i camera inside a greenhouse.

This images were inicially collected to support a robotic autonomous pollination project.

The Femoropopliteal Artery Stent Dataset is captured by Siemens digital radiography fluoroscopy system Luminos dRF Max. The dataset is from the First Affiliated Hospital of Fujian Medical University in Fuzhou City (Fujian Province, China). The X-ray fluoroscopic image data set of the femoropopliteal artery stent was obtained from different positions of the patient.

The provided dataset file contains the following directory structure when unzipped:

This publication contains three new datasets for scene-level sketch semantic segmentation task, namely SKY-Scene, TUB-Scene, and Freehand-Scene. SKY-Scene and TUB-Scene are synthetic datasets, where the scene layout templates were extracted from dataset SketchyScene, and the object components were adopted from dataset Sketchy. They are composed to include more sketch-specific characteristics, e.g., sparsity, abstractness, and diversity, to truly evaluate segmentation performance.

Fifty human-drawn scene sketches with semantic annotation.

A synthetic scene-level sketch dataset for sketch semantic segmentation task. The scene layout templates were extracted from dataset SketchyScene, and the object components were adopted from dataset TU-Berlin.

A synthetic scene-level sketch dataset for sketch semantic segmentation task. The scene layout templates were extracted from dataset SketchyScene, and the object components were adopted from dataset Sketchy.