论文阅读 【CVPR-2022】 ADAS: A Direct Adaptation Strategy for Multi-Target Domain Adaptive Semantic Segmen

论文阅读 【CVPR-2022】 ADAS: A Direct Adaptation Strategy for Multi-Target Domain Adaptive Semantic Segmentation

ADAS: 多目标领域自适应语义分割的直接适应策略

studyai.com

搜索论文: [ADAS: A Direct Adaptation Strategy for Multi-Target Domain Adaptive Semantic Segmentation](http://www.studyai.com/search/whole-site/?q=ADAS: A Direct Adaptation Strategy for Multi-Target Domain Adaptive Semantic Segmentation)

摘要(Abstract)

In this paper, we present a direct adaptation strategy (ADAS), which aims to directly adapt a single model to multiple target domains in a semantic segmentation task without pretrained domain-specific models.

在本文中，我们提出了一种直接适应策略（ADAS），其目的是在语义分割任务中直接将一个单一的模型适应于多个目标域，而不需要预先训练特定域的模型。

To do so, we design a multi-target domain transfer network (MTDT-Net) that aligns visual attributes across domains by transferring the domain distinctive features through a new target adaptive denormalization (TAD) module.

为此，我们设计了一个多目标领域迁移网络（MTDT-Net），该网络使用一个新的目标自适应去规范化（TAD）模块迁移领域的独特性特征，使各领域的视觉属性一致。

Moreover, we propose a bidirectional adaptive region selection (BARS) that reduces the attribute ambiguity among the class labels by adaptively selecting the regions with consistent feature statistics.

此外，我们提出了一个双向自适应区域选择（BARS），通过自适应地选择具有一致特征统计的区域来减少类标签之间的属性模糊性。

We show that our single MTDT-Net can synthesize visually pleasing domain transferred images with complex driving datasets, and BARS effectively filters out the unnecessary region of training images for each target domain.

我们表明，我们的单一MTDT-Net可以用复杂的驾驶数据集合成视觉上赏心悦目的领域迁移图像，而BARS可以有效地过滤掉每个目标领域训练图像中不必要的区域。

With the collaboration of MTDT-Net and BARS, our ADAS achieves state-of-the-art performance for multi-target domain adaptation (MTDA).

在MTDT-Net和BARS的合作下，我们的ADAS在多目标域适应（MTDA）方面达到了最先进的性能。

To the best of our knowledge, our method is the first MTDA method that directly adapts to multiple domains in semantic segmentation.

据我们所知，我们的方法是第一个在语义分割中直接适应多个领域的MTDA方法。