Preserving Color in Neural Artistic Style Transfer
This addresses a specific shortcoming in artistic style transfer for users who want to preserve color fidelity, but it is incremental as it builds directly on an existing method.
The paper tackles the problem of neural artistic style transfer altering colors undesirably by proposing simple linear methods to preserve colors while transferring style, resulting in a modified algorithm that maintains the original scene's appearance.
This note presents an extension to the neural artistic style transfer algorithm (Gatys et al.). The original algorithm transforms an image to have the style of another given image. For example, a photograph can be transformed to have the style of a famous painting. Here we address a potential shortcoming of the original method: the algorithm transfers the colors of the original painting, which can alter the appearance of the scene in undesirable ways. We describe simple linear methods for transferring style while preserving colors.