SkyScapes Fine-Grained Semantic Understanding of Aerial Scenes

ICCV 2019 Seyed Majid Azimi Corentin Henry Lars Sommer Arne Schumann Eleonora Vig

Understanding the complex urban infrastructure with centimeter-level accuracy is essential for many applications from autonomous driving to mapping, infrastructure monitoring, and urban management. Aerial images provide valuable information over a large area instantaneously; nevertheless, no current dataset captures the complexity of aerial scenes at the level of granularity required by real-world applications... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK LEADERBOARD
Semantic Segmentation SkyScapes-Dense SkyScapesNet-Dense Mean IoU 40.13 # 1
Semantic Segmentation SkyScapes-Lane SkyScapesNet-Lane Mean IoU 50.93 # 1