GOOSE
A multi-terrain off-road autonomous driving dataset released by the Technical University of Munich. It features high scene diversity, encompassing forests, gravel roads, paved surfaces, and complex geometric environments. The dataset is designed to support research in semantic segmentation and perception for off-road scenarios.
Timeline (last 90 days)
GOOSE (German Outdoor and Offroad Dataset) is a multi-terrain off-road dataset released by the Technical University of Munich, containing complex geometric environments such as forests, gravel roads, and paved roads.
After adding 1,500 real frames from the GOOSE dataset, sky IoU improved by 7 to 17 points, vegetation IoU improved by 29 to 37 points, and real-weather mean IoU improved by 29 to 35 points across four architectures.
Once GOOSE is included in the training set, all augmentation branches perform within ±1 point of the control group on real weather, while augmentation still adds 3.5 to 5.2 points on synthetic held-out degradations.
All Facts (3)
GOOSE (German Outdoor and Offroad Dataset) is a multi-terrain off-road dataset released by the Technical University of Munich, containing complex geometric environments such as forests, gravel roads, and paved roads.
50%UnverifiedAfter adding 1,500 real frames from the GOOSE dataset, sky IoU improved by 7 to 17 points, vegetation IoU improved by 29 to 37 points, and real-weather mean IoU improved by 29 to 35 points across four architectures.
50%UnverifiedOnce GOOSE is included in the training set, all augmentation branches perform within ±1 point of the control group on real weather, while augmentation still adds 3.5 to 5.2 points on synthetic held-out degradations.
50%