针对复杂户外场景的细粒度语义分割，提升罕见类别的识别精度。
GOOSE-M2F: Adapting Mask2Former for High-Fidelity, Long-Tailed Fine-Grained Semantic Segmentation in Unstructured Outdoor Terrain
- 增加200个对象查询，缓解特征饱和；引入双注意力模块增强特征表达。
- 通过辅助监督头和数据增广策略，使罕见类别像素级精度提升显著。
- 适合处理长尾分布的户外地形分割任务，对机器人感知有实用价值。
我们提出GOOSE-M2F，针对ICRA 2026 GOOSE 2D细粒度语义分割挑战赛的任务需求，对Mask2Former进行专项改进。该基准涵盖64个细粒度类别，分布在非结构化户外地形中，类别分布严重长尾，稀有类别每张图像不足50像素。在Swin-Large Mask2Former基础上，提出三项改进：(1) 增加200个对象查询以消除表示饱和；(2) 设计融合ASPP-lite与CBAM双注意力的特征精炼模块（FRM）；(3) 引入辅助监督头，为稀有类别提供直接像素级梯度。采用多阶段训练策略，结合分布平衡损失、稀有类复制粘贴增强、动态IoU感知重加权及EMA。推理时使用密集滑窗引擎，结合2D高斯核融合与四尺度TTA，性能提升+10.57%。最终获得70.08%官方综合mIoU（精细类63.55%，粗粒类76.61%），在排行榜中位列第3。代码与模型已开源至GitHub和Hugging Face。
原文摘要 · Abstract (English)
We present GOOSE-M2F, a task-specific adaptation of Mask2Former for the GOOSE 2D Fine-Grained Semantic Segmentation (FGSS) Challenge at ICRA 2026. The GOOSE benchmark spans 64 fine-grained classes across unstructured outdoor terrain with a severely long-tailed distribution, where rare classes occupy fewer than 50 pixels per image. We extend the Swin-Large Mask2Former baseline with three targeted contributions: (1) 200 object queries to eliminate representational saturation; (2) a Feature Refinement Module (FRM) combining ASPP-lite and CBAM dual-attention; and (3) an Auxiliary Supervision Head that delivers direct per-pixel gradients for rare classes. A multi-stage training strategy pairs Distribution-Balanced loss, Rare-Class Copy-Paste augmentation, dynamic IoU-aware re-weighting, and EMA. At inference, a dense sliding-window engine with 2D Gaussian kernel blending and 4-scale TTA adds +10.57%. GOOSE-M2F achieves 70.08% Official Composite mIoU (63.55% fine, 76.61% coarse), placing 3rd on the GOOSE 2D FGSS leaderboard. Code and trained models are publicly available at GitHub: https://github.com/Aditya-Lingam-9000/GOOSE-M2F and Hugging Face: https://huggingface.co/XYZ9843/GOOSE-M2F.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。