ELEGANT: Exchanging Latent Encodings with GAN for Transferring Multiple Face Attributes

Abstract

Recent studies on face attribute transfer have achieved great success. A lot of models are able to transfer face attributes with an input image. However, they suffer from three limitations: (1) incapability of generating image by exemplars; (2) being unable to transfer multiple face attributes simultaneously; (3) low quality of generated images, such as low-resolution or artifacts. To address these limitations, we propose a novel model which receives two images of opposite attributes as inputs. Our model can transfer exactly the same type of attributes from one image to another by exchanging certain part of their encodings. All the attributes are encoded in a disentangled manner in the latent space, which enables us to manipulate several attributes simultaneously. Besides, our model learns the residual images so as to facilitate training on higher resolution images. With the help of multi-scale discriminators for adversarial training, it can even generate high-quality images with finer details and less artifacts. We demonstrate the effectiveness of our model on overcoming the above three limitations by comparing with other methods on the CelebA face database. A pytorch implementation is available at https://github.com/Prinsphield/ELEGANT.

Github: https://github.com/Prinsphield/ELEGANT

References in corpus (8)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Toward Multimodal Image-to-Image Translation
• Invertible Conditional GANs for image editing
• Unsupervised Cross-Domain Image Generation
• Better Mixing via Deep Representations
• DNA-GAN: Learning Disentangled Representations from Multi-Attribute Images
• Modular Generative Adversarial Networks

Cited by in corpus (12)

• MaskGAN: Towards Diverse and Interactive Facial Image Manipulation
• Semantic Hierarchy Emerges in Deep Generative Representations for Scene Synthesis
• What Will Your Child Look Like? DNA-Net: Age and Gender Aware Kin Face Synthesizer
• Instance-level Facial Attributes Transfer with Geometry-Aware Flow
• StyleRig: Rigging StyleGAN for 3D Control over Portrait Images
• Texture Deformation Based Generative Adversarial Networks for Face Editing
• Sparsely Grouped Multi-task Generative Adversarial Networks for Facial Attribute Manipulation
• Domain-Specific Mappings for Generative Adversarial Style Transfer
• DLGAN: Disentangling Label-Specific Fine-Grained Features for Image Manipulation
• Shape-aware Generative Adversarial Networks for Attribute Transfer
• Bridging the Gap between Label- and Reference-based Synthesis in Multi-attribute Image-to-Image Translation
• Exemplar-based Generative Facial Editing