QuantArt:将图像风格迁移量化以实现高视觉保真度
QuantArt: Quantizing Image Style Transfer Towards High Visual Fidelity
AI总结:
针对现有风格迁移难以保证生成结果真实感的问题,本文提出 QuantArt 框架,通过向量量化及量化与连续潜在表示融合,灵活控制内容、风格与视觉保真度,并在实验中显著优于现有方法。
摘要:
现有风格迁移算法的机制是通过最小化混合损失函数,推动生成图像在内容和风格两方面都具有高度相似性。然而,这类方法无法保证视觉保真度,即生成的艺术作品应当与真实作品难以区分。本文中,我们设计了一个名为 QuantArt 的新型风格迁移框架,用于实现高视觉保真度的风格化。QuantArt 通过向量量化,将生成艺术作品的潜在表示推向真实艺术作品分布的质心。通过融合量化与连续的潜在表示,QuantArt 允许在内容保持、风格相似性和视觉保真度方面对生成艺术作品进行灵活控制。在多种风格迁移设置上的实验表明,与现有风格迁移方法相比,我们的 QuantArt 框架实现了显著更高的视觉保真度。
英文摘要:
The mechanism of existing style transfer algorithms is by minimizing a hybrid loss function to push the generated image toward high similarities in both content and style. However, this type of approach cannot guarantee visual fidelity, i.e., the generated artworks should be indistinguishable from real ones. In this paper, we devise a new style transfer framework called QuantArt for high visual-fidelity stylization. QuantArt pushes the latent representation of the generated artwork toward the centroids of the real artwork distribution with vector quantization. By fusing the quantized and continuous latent representations, QuantArt allows flexible control over the generated artworks in terms of content preservation, style similarity, and visual fidelity. Experiments on various style transfer settings show that our QuantArt framework achieves significantly higher visual fidelity compared with the existing style transfer methods.
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。