The Semantic Drone Dataset focuses on semantic understanding of urban scenes for increasing the safety of autonomous drone flight and landing procedures.
The imagery depicts more than 20 houses from bird's eye view acquired at an altitude of 5 to 30 meters above ground. A high resolution camera was used to acquire images at a size of 6000x4000px (24Mpx).
The training set contains 400 publicly available images and the test set is made up of 200 private images.
No associated paper or article has been found in regard to this dataset.
Person Detection: For the task of person detection the dataset contains bounding box annotations of the training and test set.
Semantic Segmentation: We prepared pixel-accurate annotation for the same training and test set. The complexity of the dataset is limited to 20 classes as listed in the following table:
|
|
|
|
Sample clean and annotated image:
To download the dataset you ought to fill out the following google form.
No associated model has been provided.
No benchmarks have been provided.
No assiciated challenges have been found in regard to this dataset.
The dataset is licenced under a non-commercial licence. More details can be found on the offical webpage.