轻量高效去雾网络，小数据也能精准去雾。
DRACO-DehazeNet: An Efficient Image Dehazing Network Combining Detail Recovery and a Novel Contrastive Learning Paradigm
- 用密集空洞残差块+注意力细节恢复，自适应提升不同场景去雾效果。
- 仅需少量数据训练，通过四元组对比损失提升去雾质量。
- 适合移动端部署，尤其在重雾或不均匀雾霾下表现更优。
图像去雾对改善雾霾遮挡的模糊图像至关重要，但现有基于学习的方法依赖大量训练数据，计算开销大，且在非均匀或重度雾霾下性能不佳。为此，我们提出细节恢复与对比去雾网络（DRACO-DehazeNet），通过密集空洞反向残差块和基于注意力的细节恢复网络，针对特定去雾场景进行优化。其核心创新在于采用基于四元组损失的对比去雾范式，有效分离有雾与无雾图像特征，并区分各子模块生成的不同质量去雾结果，从而显著提升去雾精度。在多个基准雾霾数据集上的实验表明，该方法具有明显优势。代码已开源：https://github.com/GreedYLearner1146/DRACO-DehazeNet。
原文摘要 · Abstract (English)
Image dehazing is crucial for clarifying images obscured by haze or fog, but current learning-based approaches is dependent on large volumes of training data and hence consumed significant computational power. Additionally, their performance is often inadequate under non-uniform or heavy haze. To address these challenges, we developed the Detail Recovery And Contrastive DehazeNet, which facilitates efficient and effective dehazing via a dense dilated inverted residual block and an attention-based detail recovery network that tailors enhancements to specific dehazed scene contexts. A major innovation is its ability to train effectively with limited data, achieved through a novel quadruplet loss-based contrastive dehazing paradigm. This approach distinctly separates hazy and clear image features while also distinguish lower-quality and higher-quality dehazed images obtained from each sub-modules of our network, thereby refining the dehazing process to a larger extent. Extensive tests on a variety of benchmarked haze datasets demonstrated the superiority of our approach. The code repository for this work is available at https://github.com/GreedYLearner1146/DRACO-DehazeNet.
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。