Two-phase Hair Image Synthesis by Self-Enhancing Generative Model
This work addresses the challenge of realistic hair synthesis for computer graphics and image processing applications, representing an incremental improvement over existing GAN-based methods.
The paper tackles the problem of generating high-quality hair images from sparse sketches or low-resolution inputs by proposing a two-phase generative model that first creates a coarse image and then refines it with a self-enhancing network, achieving state-of-the-art results with finer details in tasks like Sketch2Hair and Hair Super-Resolution.
Generating plausible hair image given limited guidance, such as sparse sketches or low-resolution image, has been made possible with the rise of Generative Adversarial Networks (GANs). Traditional image-to-image translation networks can generate recognizable results, but finer textures are usually lost and blur artifacts commonly exist. In this paper, we propose a two-phase generative model for high-quality hair image synthesis. The two-phase pipeline first generates a coarse image by an existing image translation model, then applies a re-generating network with self-enhancing capability to the coarse image. The self-enhancing capability is achieved by a proposed structure extraction layer, which extracts the texture and orientation map from a hair image. Extensive experiments on two tasks, Sketch2Hair and Hair Super-Resolution, demonstrate that our approach is able to synthesize plausible hair image with finer details, and outperforms the state-of-the-art.