Two-phase Hair Image Synthesis by Self-Enhancing Generative Model
arXiv:1902.11203 · doi:10.1111/cgf.13847
Abstract
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.
References in corpus (11)
- Conditional Generative Adversarial Nets
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- Conditional Random Fields as Recurrent Neural Networks
- Progressive Growing of GANs for Improved Quality, Stability, and Variation
- Photographic Image Synthesis with Cascaded Refinement Networks
- Look, Listen and Learn - A Multimodal LSTM for Speaker Identification
- Deep Cascaded Bi-Network for Face Hallucination
- Learning Dual Convolutional Neural Networks for Low-Level Vision
- MGANet: A Robust Model for Quality Enhancement of Compressed Video
- FaceShop: Deep Sketch-based Face Image Editing
- HairNet: Single-View Hair Reconstruction using Convolutional Neural Networks