Technical Report for ICRA 2026 GOOSE 2D Fine-Grained Semantic Segmentation Challenge: Exploring Query-Based Segmentation and Increased Spatial Context for Outdoor Scene Understanding

2026-06-19

AI Summary

This work addresses the challenge of fine-grained semantic segmentation in outdoor scenes, where semantic ambiguity arises from highly granular class distinctions and scene complexity. Building upon SegFormer as a baseline, the study introduces a query-based Mask2Former architecture augmented with large-scale training crops and test-time augmentation strategies to significantly enhance spatial context modeling and fine-grained discriminative capability. Experimental results demonstrate that increasing input resolution is crucial for preserving semantic details, and that query-based segmentation frameworks offer distinct advantages in complex outdoor environments. The proposed method achieves a mean Intersection-over-Union (mIoU) of 69.6% on the GOOSE challenge test set, establishing a strong baseline for fine-grained outdoor semantic segmentation.

Abstract

In this report, we present our submission to the GOOSE 2D Fine-Grained Semantic Segmentation Challenge, organized as part of the Workshop on Field Robotics at ICRA 2026. The challenge combines data from the GOOSE and GOOSE-Ex datasets, which comprise more than 13k images captured from 4 distinct camera setups, annotated using a hierarchical taxonomy of 56 fine-grained classes and 11 broader categories. Starting from SegFormer as a baseline, we progressively improve segmentation performance through increased training crop sizes, a transition to the query-based Mask2Former architecture, and test-time augmentation. Our experiments show that query-based segmentation significantly outperforms the baseline model. Furthermore, increasing the crop size used during training yields substantial gains, highlighting the relevance of preserving scene context for fine-grained semantic disambiguation. Our final submission, using test-time augmentation, achieves an mIoU of 69.6% on the challenge test set, providing a strong baseline for fine-grained semantic segmentation in outdoor environments. To facilitate reproducibility and future research, code and weights will be made publicly available at https://github.com/RoboticsLabURJC/outdoor-fine-grained-segmentation .

Problem

• fine-grained semantic segmentation
• outdoor scene understanding
• semantic disambiguation
• hierarchical taxonomy
• 2D segmentation

Innovation

• query-based segmentation
• spatial context
• fine-grained semantic segmentation
• Mask2Former
• test-time augmentation