Due to the high diversity of image styles, the scalability to various styles plays a critical role in real-world applications. To accommodate a large amount of styles, previous multi-style transfer approaches rely on enlarging the model size while arbitrary-style transfer methods utilize heavy backbones. However, the additional computational cost introduced by more model parameters hinders these methods to be deployed on resource-limited devices. To address this challenge, in this paper, we develop a style transfer framework by decoupling the style modeling and transferring. Specifically, for style modeling, we propose a style representation learning scheme to encode the style information into a compact representation. Then, for style transferring, we develop a style-aware multi-style transfer network (SaMST) to adapt to diverse styles using pluggable style representations. In this way, our framework is able to accommodate diverse image styles in the learned style representations without introducing additional overhead during inference, thereby maintaining efficiency. Experiments show that our style representation can extract accurate style information. Moreover, qualitative and quantitative results demonstrate that our method achieves state-of-the-art performance in terms of both accuracy and efficiency. The codes are available in https://github.com/The-Learning-And-Vision-Atelier-LAVA/SaMST.

翻译：由于图像风格的高度多样性，对各种风格的可扩展性在实际应用中起着关键作用。为适应大量风格，以往的多风格迁移方法依赖于扩大模型规模，而任意风格迁移方法则使用重型骨干网络。然而，更多模型参数带来的额外计算成本阻碍了这些方法在资源受限设备上的部署。为应对这一挑战，本文通过解耦风格建模与风格迁移，构建了一个风格迁移框架。具体而言，在风格建模方面，我们提出一种风格表征学习方案，将风格信息编码为紧凑的表征。在风格迁移方面，我们开发了一种风格感知多风格迁移网络（SaMST），通过可插拔的风格表征来适应多样风格。通过这种方式，我们的框架能够在学习到的风格表征中容纳多样化的图像风格，且在推理过程中不引入额外开销，从而保持高效性。实验表明，我们的风格表征能够提取准确的风格信息。此外，定性与定量结果均证明，我们的方法在准确性和效率方面均达到了最先进的性能。代码发布于 https://github.com/The-Learning-And-Vision-Atelier-LAVA/SaMST。