2D Fine-Grained Semantic Segmentation

My Logo

The GOOSE 2D Fine-Grained Semantic Segmentation Challenge was a competition hosted in conjunction with the Workshop on Field Robotics at ICRA 2026 in Vienna, Austria. The challenge is based on the annotated image data of unstructured outdoor environments found in the GOOSE and GOOSE-Ex datasets.

Participants were tasked with developing and testing models for 2D Semantic Segmentation of camera images taken from three different robotic platforms. The teams with the best-performing models were awarded prizes and the opportunity to present their approach during the poster session of the Workshop on Field Robotics at ICRA 2026.

The classes and category segmentation are based on the GOOSE Class Definitions.

GOOSE Annotation

2026 Finalists

  1. 馃 MIP Lab (DGIST x MIT) with a mIoU of 76.57% 路 [technical report]
  2. 馃 HFUT-LION with a mIoU of 75.40% 路 [technical report]
  3. 馃 NJyothiraditya Lingam, Nikhileswara Rao Sulake and Sai Manikanta Eswar Machara with a mIoU of 70.08% 路 [technical report] 路 [code] 路 [model weights]
  4. Xuesong Wang with a mIoU of 69.73% 路 [technical report]
  5. David Pascual-Hern谩ndez, Roberto Calvo-Palomino, Inmaculada Mora-Jim茅nez and Jose Mar铆a Ca帽as-Plaza with a mIoU of 69.62% 路 [[technical report]