arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.13447cs.CV
收藏

DRANet:用于无监督跨领域自适应的解纠缠表示和自适应网络

DRANet: Disentangling Representation and Adaptation Networks for Unsupervised Cross-Domain Adaptation

DGIST(大邱庆北科学技术院);POSTECH(浦项科技大学)

Seunghun Lee, Sunghyun Cho, Sunghoon Im

摘要:

在本文中,我们提出了DRANet,一个网络架构,解开图像表示和转移的视觉属性在一个潜在的空间,无监督的跨领域自适应。与现有的领域自适应方法,学习共享域的相关功能,DRANet保留了每个域的特性的独特性。我们的模型编码的内容(场景结构)和风格(艺术外观)从源和目标图像的个人表示。然后,它通过将传输的风格因子与为每个域指定的可学习权重一起合并到内容因子中来适应域。该学习框架允许利用单个编码器-解码器网络进行双向/多向领域自适应,并对齐它们的域移位。此外,我们提出了一个内容自适应域转移模块,有助于保留场景结构,同时转移风格。大量的实验表明,我们的模型成功地分离内容风格的因素,并合成视觉上令人愉快的域转移图像。所提出的方法在标准数字分类任务以及语义分割任务上展示了最先进的性能。

英文摘要:

In this paper, we present DRANet, a network architecture that disentangles image representations and transfers the visual attributes in a latent space for unsupervised cross-domain adaptation. Unlike the existing domain adaptation methods that learn associated features sharing a domain, DRANet preserves the distinctiveness of each domain's characteristics. Our model encodes individual representations of content (scene structure) and style (artistic appearance) from both source and target images. Then, it adapts the domain by incorporating the transferred style factor into the content factor along with learnable weights specified for each domain. This learning framework allows bi-/multi-directional domain adaptation with a single encoder-decoder network and aligns their domain shift. Additionally, we propose a content-adaptive domain transfer module that helps retain scene structure while transferring style. Extensive experiments show our model successfully separates content-style factors and synthesizes visually pleasing domain-transferred images. The proposed method demonstrates state-of-the-art performance on standard digit classification tasks as well as semantic segmentation tasks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑