MUAD

We introduce MUAD, a synthetic dataset for autonomous driving with multiple uncertainty types and tasks. It contains 10413 in total: 3420 images in the train set, 492 in the validation set and 6501 in the test set. The test set is divided as follows: 551 in the normal set, 102 in the normal set no shadow, 1668 in the OOD set, 605 in the low adversity set and 602 images in the high adversity set 1552 in the low adversity with OOD set and 1421 images in the high adversity with OOD set. All of these sets cover day and night conditions, with 2/3 being day images and 1/3 night images. Test datasets address diverse weather conditions (rain, snow, and fog with two different intensity levels) and multiple OOD objects.

MUAD has seven test sets:

• normal set: images without OOD objects nor adverse conditions
• low adversity with OOD set: images containing both OOD objects and medium intensity adversity conditions (either fog, rain or snow)
• high adversity with OOD set: images containing both OOD objects and high intensity adversity conditions (either fog, rain or snow)
• normal set no shadow: images without OOD objects nor adversity conditions and we simulate the sun like if it was 12:00 am so the shadow is minimum
• OOD set: images containing OOD objects and without adversity conditions
• low adversity set: images containing medium intensity adversity conditions (either fog, rain or snow)
• high adversity set: images containing high intensity adversity conditions (either fog, rain or snow)

MUAD supports four tasks:

• Semantic segmentation
• Depth estimation
• Object detection
• Instance detection

Overview of annotated classes:

There are a total of 155 fine grained classes, which are also aggregated in order to facilitate the use along with other datasets, e.g. Citiscapes: