Deep neural networks for scene perception in automated vehicles achieve excellent results for the domains they were trained on. However, in real-world conditions, the domain of operation and its underlying data distribution are subject to change. Adverse weather conditions, in particular, can significantly decrease model performance when such data are not available during training.Additionally, when a model is incrementally adapted to a new domain, it suffers from catastrophic forgetting, causing a significant drop in performance on previously observed domains. Despite recent progress in reducing catastrophic forgetting, its causes and effects remain obscure. Therefore, we study how the representations of semantic segmentation models are affected during domain-incremental learning in adverse weather conditions. Our experiments and representational analyses indicate that catastrophic forgetting is primarily caused by changes to low-level features in domain-incremental learning and that learning more general features on the source domain using pre-training and image augmentations leads to efficient feature reuse in subsequent tasks, which drastically reduces catastrophic forgetting. These findings highlight the importance of methods that facilitate generalized features for effective continual learning algorithms.

翻译：面向自动驾驶场景感知的深度神经网络在训练所覆盖的领域上取得了优异表现。然而在实际工况中，运行领域及其底层数据分布会发生变化，特别是恶劣天气条件——当此类数据在训练阶段缺失时，会显著降低模型性能。此外，当模型逐步适应新领域时，会遭遇灾难性遗忘现象，导致其在先前观测领域上的性能大幅下降。尽管近年来在缓解灾难性遗忘方面取得了进展，但其成因与影响机制仍不明确。为此，本文研究了恶劣天气条件下语义分割模型在领域增量学习中的表征演化规律。实验与表征分析表明：领域增量学习中的灾难性遗忘主要由底层特征变化引发；通过在源领域采用预训练与图像增强技术学习更通用的特征，可促使后续任务中的高效特征复用，从而显著降低灾难性遗忘程度。这些发现揭示了构建促进特征泛化方法对于设计高效持续学习算法具有重要价值。