论文标题
归因感知体重传递：班级插图的温暖初始化
Attribution-aware Weight Transfer: A Warm-Start Initialization for Class-Incremental Semantic Segmentation
论文作者
论文摘要
在课堂开发语义细分（CISS）中，深度学习体系结构遭受了灾难性遗忘和语义背景转移的关键问题。尽管最近的工作着重于这些问题，但现有的分类器初始化方法并未解决背景转移问题，并将相同的初始化权重分配给背景和新的前景类别分类器。我们建议使用一种新型分类器初始化方法来解决背景转移，该方法采用基于梯度的归属来确定从分类器的权重以先前背景的权重，并将这些权重传递给新分类器。这种温暖的重量初始化提供了适用于几种CISS方法的通用解决方案。此外，它在减轻遗忘的同时加快了学习新课程的学习。我们的实验表明，与Pascal-Voc 2012，ADE20K和CityScapes数据集的最新CISS方法相比，MIOU显着改善。
In class-incremental semantic segmentation (CISS), deep learning architectures suffer from the critical problems of catastrophic forgetting and semantic background shift. Although recent works focused on these issues, existing classifier initialization methods do not address the background shift problem and assign the same initialization weights to both background and new foreground class classifiers. We propose to address the background shift with a novel classifier initialization method which employs gradient-based attribution to identify the most relevant weights for new classes from the classifier's weights for the previous background and transfers these weights to the new classifier. This warm-start weight initialization provides a general solution applicable to several CISS methods. Furthermore, it accelerates learning of new classes while mitigating forgetting. Our experiments demonstrate significant improvement in mIoU compared to the state-of-the-art CISS methods on the Pascal-VOC 2012, ADE20K and Cityscapes datasets.
