Line Search-Based Feature Transformation for Fast, Stable, and Tunable Content-Style Control in Photorealistic Style Transfer
This work addresses the need for efficient and tunable style transfer in image processing, though it appears incremental as it builds on existing models.
The paper tackles photorealistic style transfer by introducing a feature transformation that balances content preservation and style strength, achieving faster runtime and more consistent results compared to existing methods.
Photorealistic style transfer is the task of synthesizing a realistic-looking image when adapting the content from one image to appear in the style of another image. Modern models commonly embed a transformation that fuses features describing the content image and style image and then decodes the resulting feature into a stylized image. We introduce a general-purpose transformation that enables controlling the balance between how much content is preserved and the strength of the infused style. We offer the first experiments that demonstrate the performance of existing transformations across different style transfer models and demonstrate how our transformation performs better in its ability to simultaneously run fast, produce consistently reasonable results, and control the balance between content and style in different models. To support reproducing our method and models, we share the code at https://github.com/chiutaiyin/LS-FT.