CAP-VSTNet: Content Affinity Preserved Versatile Style Transfer
This addresses artifacts in style transfer for applications like image and video editing, but appears incremental as it builds on existing reversible network and linear transform ideas.
The paper tackles the problem of content affinity loss causing artifacts in photorealistic and video style transfer by proposing CAP-VSTNet, a framework with a reversible residual network and unbiased linear transform module, which achieves better qualitative and quantitative results compared to state-of-the-art methods.
Content affinity loss including feature and pixel affinity is a main problem which leads to artifacts in photorealistic and video style transfer. This paper proposes a new framework named CAP-VSTNet, which consists of a new reversible residual network and an unbiased linear transform module, for versatile style transfer. This reversible residual network can not only preserve content affinity but not introduce redundant information as traditional reversible networks, and hence facilitate better stylization. Empowered by Matting Laplacian training loss which can address the pixel affinity loss problem led by the linear transform, the proposed framework is applicable and effective on versatile style transfer. Extensive experiments show that CAP-VSTNet can produce better qualitative and quantitative results in comparison with the state-of-the-art methods.