26 Cool Cityscapes

Cityscapes is a dataset of images and annotations for urban street scenes. It was created for use in computer vision research, particularly for the task of semantic urban scene understanding. The dataset includes a large collection of high-resolution images of street scenes from different cities, along with finely-annotated labels for various objects and features in the images, such as buildings, roads, pedestrians, and vehicles. The annotations include both object-level labels (e.g. “car”, “person”) and instance-level labels (e.g. “car 1”, “car 2”). The Cityscapes dataset is commonly used for semantic segmentation tasks and object detection in autonomous driving.