Precise Recovery of Latent Vectors from Generative Adversarial Networks

Abstract

Generative adversarial networks (GANs) transform latent vectors into visually plausible images. It is generally thought that the original GAN formulation gives no out-of-the-box method to reverse the mapping, projecting images back into latent space. We introduce a simple, gradient-based technique called stochastic clipping. In experiments, for images generated by the GAN, we precisely recover their latent vector pre-images 100% of the time. Additional experiments demonstrate that this method is robust to noise. Finally, we show that even for unseen images, our method appears to recover unique encodings.

References in corpus (3)

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Adversarial Feature Learning
• Inverting The Generator Of A Generative Adversarial Network (II)

Cited by in corpus (28)

• Optimizing the Latent Space of Generative Networks
• Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis
• Gradient-based deterministic inversion of geophysical data with Generative Adversarial Networks: is it feasible?
• Semi-supervised Image Attribute Editing using Generative Adversarial Networks
• Generative adversarial networks and adversarial methods in biomedical image analysis
• Unsupervised Image-to-Image Translation via Pre-trained StyleGAN2 Network
• Comparing Generative Adversarial Network Techniques for Image Creation and Modification
• Semantic Segmentation with Generative Models: Semi-Supervised Learning and Strong Out-of-Domain Generalization
• Unsupervised 3D Shape Completion through GAN Inversion
• Improving ClusterGAN Using Self-Augmented Information Maximization of Disentangling Latent Spaces
• Learning a low dimensional manifold of real cancer tissue with PathologyGAN
• Towards Recovery of Conditional Vectors from Conditional Generative Adversarial Networks
• SyncGAN: Synchronize the Latent Space of Cross-modal Generative Adversarial Networks
• Learning Formation of Physically-Based Face Attributes
• Latent Variable Modeling for Generative Concept Representations and Deep Generative Models
• Deep Meditations: Controlled navigation of latent space
• GAN-Based Facial Attractiveness Enhancement
• Bayesian Reasoning with Trained Neural Networks
• Exploiting GAN Internal Capacity for High-Quality Reconstruction of Natural Images
• Latent Vector Recovery of Audio GANs
• Neural Crossbreed: Neural Based Image Metamorphosis
• Inverse mapping of face GANs
• Robust Training Using Natural Transformation
• Matched sample selection with GANs for mitigating attribute confounding
• Ensembles of Generative Adversarial Networks for Disconnected Data
• The Effects of Invertibility on the Representational Complexity of Encoders in Variational Autoencoders
• Force-in-domain GAN inversion
• Directional GAN: A Novel Conditioning Strategy for Generative Networks