用训练技巧弥补恶劣天气分割的泛化差距，效果优于堆模型。
Bridging the Generalization Gap in Adverse Weather Segmentation: A Training Recipe Perspective
- 通过数据混合、场景平衡采样等训练策略提升鲁棒性。
- 测试集mIoU达59.9%，验证与测试差距仅6.5点。
- 适合在数据有限时构建抗恶劣天气的分割系统。
本文介绍我们在2026年CVPR UG2+ Workshop Track 2中的方法，针对五种恶劣天气（模糊、黑暗、雪、雾霾、强光）下的室外场景语义分割挑战。观察到的核心问题是严重泛化差距——模型在验证集表现良好，但在测试集上性能急剧下降，例如SegFormer-B5从验证到测试损失16.1 mIoU，表明模型容量不足以保证鲁棒性。我们探究是否可通过精心设计的训练配方而非架构复杂度来缓解该问题。基于预训练的SegMAN-S主干网络，系统研究了领域自适应微调、多源数据混合、场景平衡采样及合成退化增强的影响。最终系统在官方测试集上达到59.9% mIoU，且验证-测试差距仅为6.5点，不足大型模型的一半。通过分析架构修改、损失函数变体和模型缩放的负面结果，为有限数据下实现天气鲁棒分割提供了实用洞见。
原文摘要 · Abstract (English)
This paper describes our approach for the 8th UG2+ Workshop (CVPR 2026) Track~2, which targets semantic segmentation of outdoor scenes degraded by five weather conditions: blur, darkness, snow, haze, and glare. A central challenge we observe is a severe generalization gap -- models that perform well on the validation set often collapse on the test set. For instance, SegFormer-B5 drops 16.1 mIoU points from validation to test, suggesting that model capacity alone is insufficient for robustness. We investigate whether a carefully designed training recipe, rather than architectural complexity, can address this gap. Starting from a pre-trained SegMAN-S backbone, we systematically study the effects of domain-adaptive fine-tuning, multi-source data mixing, scene-balanced sampling, and synthetic degradation augmentation. Our final system achieves 59.9\% mIoU on the official test set while maintaining a validation-test gap of only 6.5 points -- less than half that of larger models. We analyze negative results from architectural modifications, loss function variants, and model scaling to provide practical insights for weather-robust segmentation under limited data.
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。