改进局部风格迁移，精准只对目标区域上色。
Improving Masked Style Transfer using Blended Partial Convolution
- 用部分卷积网络聚焦目标区域，避免风格泄露。
- 在SA-1B数据集上，视觉与定量指标均显著提升。
- 适合需要精细控制风格迁移范围的研究者。
艺术风格迁移得益于卷积与基于Transformer的神经网络发展。现有方法通常对整张图像应用风格，但用户可能仅需对图像特定区域进行处理。传统做法是在风格化后加掩码，但该方法容易在关注区域错误捕获风格特征。本文提出一种基于部分卷积的风格迁移网络，能将风格特征精确地仅应用于目标区域。同时引入网络内部的融合技术，缓解区域选择不完美的问题。实验基于SA-1B数据集，验证了该方法在视觉效果和量化指标上的优势。代码已公开于https://github.com/davidmhart/StyleTransferMasked。
原文摘要 · Abstract (English)
Artistic style transfer has long been possible with the advancements of convolution- and transformer-based neural networks. Most algorithms apply the artistic style transfer to the whole image, but individual users may only need to apply a style transfer to a specific region in the image. The standard practice is to simply mask the image after the stylization. This work shows that this approach tends to improperly capture the style features in the region of interest. We propose a partial-convolution-based style transfer network that accurately applies the style features exclusively to the region of interest. Additionally, we present network-internal blending techniques that account for imperfections in the region selection. We show that this visually and quantitatively improves stylization using examples from the SA-1B dataset. Code is publicly available at https://github.com/davidmhart/StyleTransferMasked.
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。