让模型自适应恶劣天气，仅用少量训练就显著提升检测精度。
WARLearn: Weather-Adaptive Representation Learning
- 基于不变性原理，迁移已有清晰图像模型应对雾霾和低光。
- 在真实雾天数据集上达到52.6% mAP，低光下达55.7% mAP。
- 适合需要持续更新的动态环境，尤其适用于天气变化大的场景。
本文提出WARLearn，一种面向恶劣与对抗性天气条件的自适应表征学习框架。利用Barlow Twins中的不变性原则，该方法可将原本在晴朗天气数据上训练的模型，高效迁移至处理雾霾与低光等复杂场景。仅需少量额外训练，便在雾天与低光条件下展现出显著性能提升。实验表明，在未见过的真实雾天数据集RTTS上，mAP达52.6%；在低光数据集ExDark上，mAP达55.7%。相比FeatEnHancer、Image Adaptive YOLO、DENet、C2PNet、PairLIE及ZeroDCE等前沿框架，本方法在恶劣天气下表现更优，大幅超越基线。WARLearn代码已开源。
原文摘要 · Abstract (English)
This paper introduces WARLearn, a novel framework designed for adaptive representation learning in challenging and adversarial weather conditions. Leveraging the in-variance principal used in Barlow Twins, we demonstrate the capability to port the existing models initially trained on clear weather data to effectively handle adverse weather conditions. With minimal additional training, our method exhibits remarkable performance gains in scenarios characterized by fog and low-light conditions. This adaptive framework extends its applicability beyond adverse weather settings, offering a versatile solution for domains exhibiting variations in data distributions. Furthermore, WARLearn is invaluable in scenarios where data distributions undergo significant shifts over time, enabling models to remain updated and accurate. Our experimental findings reveal a remarkable performance, with a mean average precision (mAP) of 52.6% on unseen real-world foggy dataset (RTTS). Similarly, in low light conditions, our framework achieves a mAP of 55.7% on unseen real-world low light dataset (ExDark). Notably, WARLearn surpasses the performance of state-of-the-art frameworks including FeatEnHancer, Image Adaptive YOLO, DENet, C2PNet, PairLIE and ZeroDCE, by a substantial margin in adverse weather, improving the baseline performance in both foggy and low light conditions. The WARLearn code is available at https://github.com/ShubhamAgarwal12/WARLearn
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。