Outdoor Semantic Segmentation Challenge Call for Papers
Outdoor Semantic Segmentation Challenge @ DAGM GCPR 2021: Call for Participation
Machine Learning Competition on an Outdoor Driving Dataset
(TAS500v1.1) with a submission deadline on August 15, 2021. The best
results are presented as part of the SUUE workshop @ DAGM GCPR 2021 on
September 28, 2021 in Bonn, Germany.
* August 15, 2021 - Challenge Deadline
* September 28, 2021 - Workshop Presentation Date
The vast majority of research in scene understanding is applied in
urban and structured environments. Unfortunately, these methods often
fail when used in unstructured environments with challenging lighting
conditions and many natural structures in the scene.
We therefore invite you to participate in the Outdoor Semantic
Segmentation Challenge and give you the opportunity to benchmark your
semantic segmentation algorithms on a novel dataset of finely-grained
image annotations of real world outdoor driving scenes.
The challenge uses the TAS500 dataset, which consists of over 500
annotated images of different outdoor environments. The dataset
applies a fine-grained vegetation and road surface annotation
policy. The segmented scenes are of interest for applications in
forestry robotics or autonomous driving in unstructured environments.
The challenge is part of the 1st Workshop on Scene Understanding in
Unstructured Evironments (SUUE 2021) and is organized in conjunction
with the DAGM GCPR 2021. The results on the test set will be published
on the public leaderboard on Codalab. The participants with the best
submissions will be invited to present their results at the SUUE
workshop on September 28, 2021.
The workshop presentation can also be pre-recorded or done
remotely. We therefore invite the international research community to
participate in the competition.
Peter Mortimer (Bundeswehr University Munich)
For any additional information feel free to contact us at
We are looking forward to your competition submissions and your
attendance at the SUUE 2021 this September!