Example-Guided Style Consistent Image Synthesis from Semantic Labeling
This work addresses the challenge of generating realistic images with consistent style for applications like portrait or scene synthesis, representing an incremental improvement over existing methods.
The paper tackles the problem of synthesizing images from semantic label maps guided by an example image to ensure style consistency, proposing a conditional GAN with a style consistency discriminator, adaptive semantic consistency loss, and training data sampling strategy, achieving improved results in generating style-consistent images.
Example-guided image synthesis aims to synthesize an image from a semantic label map and an exemplary image indicating style. We use the term "style" in this problem to refer to implicit characteristics of images, for example: in portraits "style" includes gender, racial identity, age, hairstyle; in full body pictures it includes clothing; in street scenes, it refers to weather and time of day and such like. A semantic label map in these cases indicates facial expression, full body pose, or scene segmentation. We propose a solution to the example-guided image synthesis problem using conditional generative adversarial networks with style consistency. Our key contributions are (i) a novel style consistency discriminator to determine whether a pair of images are consistent in style; (ii) an adaptive semantic consistency loss; and (iii) a training data sampling strategy, for synthesizing style-consistent results to the exemplar.