Semi-Latent GAN: Learning to generate and modify facial images from attributes

Abstract

Generating and manipulating human facial images using high-level attributal controls are important and interesting problems. The models proposed in previous work can solve one of these two problems (generation or manipulation), but not both coherently. This paper proposes a novel model that learns how to both generate and modify the facial image from high-level semantic attributes. Our key idea is to formulate a Semi-Latent Facial Attribute Space (SL-FAS) to systematically learn relationship between user-defined and latent attributes, as well as between those attributes and RGB imagery. As part of this newly formulated space, we propose a new model --- SL-GAN which is a specific form of Generative Adversarial Network. Finally, we present an iterative training algorithm for SL-GAN. The experiments on recent CelebA and CASIA-WebFace datasets validate the effectiveness of our proposed framework. We will also make data, pre-trained models and code available.

10 pages, submitted to ICCV 2017

References in corpus (7)

• Deep Learning Face Representation by Joint Identification-Verification
• Learning Face Representation from Scratch
• Deep Convolutional Inverse Graphics Network
• Invertible Conditional GANs for image editing
• Disentangling factors of variation in deep representations using adversarial training
• Learning What and Where to Draw
• Plug & Play Generative Networks: Conditional Iterative Generation of Images in Latent Space

Cited by in corpus (13)

• How Generative Adversarial Networks and Their Variants Work: An Overview
• A Review on Generative Adversarial Networks: Algorithms, Theory, and Applications
• Face Recognition: From Traditional to Deep Learning Methods
• Semantically Decomposing the Latent Spaces of Generative Adversarial Networks
• Pose-Normalized Image Generation for Person Re-identification
• Speech-Driven Facial Reenactment Using Conditional Generative Adversarial Networks
• Controllable Person Image Synthesis with Attribute-Decomposed GAN
• Recent Progress of Face Image Synthesis
• Deep Generative Modeling-based Data Augmentation with Demonstration using the BFBT Benchmark Void Fraction Datasets
• A Survey and Taxonomy of Adversarial Neural Networks for Text-to-Image Synthesis
• A Large-scale Attribute Dataset for Zero-shot Learning
• Adjusting Decision Boundary for Class Imbalanced Learning
• Editable Generative Adversarial Networks: Generating and Editing Faces Simultaneously