SWIMSEG
Singapore Whole sky IMaging SEGmentation Database
Creative Commons LicenseIntroduced 2016-06-12
The SWIMSEG dataset contains 1013 images of sky/cloud patches, along with their corresponding binary segmentation maps. The ground truth annotation was done in consultation with experts from Singapore Meteorological Services. All images were captured in Singapore using WAHRSIS, a calibrated ground-based whole sky imager, over a period of 22 months from October 2013 to July 2015. Each patch covers about 60-70 degrees of the sky with a resolution of 600x600 pixels.
Source: Color-based segmentation of sky/cloud images from ground-based cameras
Benchmarks
10-shot image generation/Average Precision10-shot image generation/Average Recall10-shot image generation/F1-Score10-shot image generation/MCC10-shot image generation/Mean IoUSemantic Segmentation/Average PrecisionSemantic Segmentation/Average RecallSemantic Segmentation/F1-ScoreSemantic Segmentation/MCCSemantic Segmentation/Mean IoU