GANSpace: Discovering Interpretable GAN Controls

2020-04-06 · NeurIPS 2020 12 · Erik Härkönen, Aaron Hertzmann, Jaakko Lehtinen, Sylvain Paris

This paper describes a simple technique to analyze Generative Adversarial Networks (GANs) and create interpretable controls for image synthesis, such as change of viewpoint, aging, lighting, and time of day. We identify important latent directions based on Principal Components Analysis (PCA) applied either in latent space or feature space. Then, we show that a large number of interpretable controls can be defined by layer-wise perturbation along the principal directions. Moreover, we show that BigGAN can be controlled with layer-wise inputs in a StyleGAN-like manner. We show results on different GANs trained on various datasets, and demonstrate good qualitative matches to edit directions found through earlier supervised approaches.

📄 PDF Abstract BibTeX arXiv:2004.02546

Code (2)

harskish/ganspace 공식 구현 pytorch

woctezuma/steam-stylegan2 pytorch

Tasks

Image Generation

Methods 이 논문이 사용한 방법론

Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

ReLU How Do I Communicate to Expedia? How Do I Communicate to Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Live Support & Special Travel…

Feedforward Network A Feedforward Network, or a Multilayer Perceptron (MLP), is a neural network with solely densely connected layers. This is the classic neural network architecture of the…

Conditional Batch Normalization Conditional Batch Normalization (CBN) is a class-conditional variant of batch normalization. The key idea is to…

Residual Block Residual Blocks are skip-connection blocks that learn residual functions with reference to the layer inputs, instead of learning unreferenced functions. They were introduced…

TTUR The Two Time-scale Update Rule (TTUR) is an update rule for generative adversarial networks trained with stochastic gradient descent. TTUR has an individual learning rate for…

GAN Hinge Loss The GAN Hinge Loss is a hinge loss based loss function for [generative adversarial…

WarpedGANSpace: Finding non-linear RBF paths in GAN latent space

2021-09-27 · ICCV 2021 10 · Christos Tzelepis, Georgios Tzimiropoulos, Ioannis Patras

This work addresses the problem of discovering, in an unsupervised manner, interpretable paths in the latent space of pretrained GANs, so as to provide an intuitive and easy way of controlling the underlying generative f…

Exploring Semantic Variations in GAN Latent Spaces via Matrix Factorization

2023-05-23 · Andrey Palaev, Rustam A. Lukmanov, Adil Khan

Controlled data generation with GANs is desirable but challenging due to the nonlinearity and high dimensionality of their latent spaces. In this work, we explore image manipulations learned by GANSpace, a state-of-the-a…

Disentanglement

Automatic Discovery of Disease Subgroups by Contrasting with Healthy Controls

2026-05-20 · Robin Louiset, Edouard Duchesnay, Benoit Dufumier, Antoine Grigis 외 arxiv

In biomedical Subgroup Discovery, practitioners are interested in discovering interpretable and homogeneous subgroups within a group of patients. In this paper, assuming that healthy subjects (i.e., controls) share commo…

Discovering Class-Specific GAN Controls for Semantic Image Synthesis

2022-12-02 · Edgar Schönfeld, Julio Borges, Vadim Sushko, Bernt Schiele 외

Prior work has extensively studied the latent space structure of GANs for unconditional image synthesis, enabling global editing of generated images by the unsupervised discovery of interpretable latent directions. Howev…

Image Generation

DragGANSpace: Latent Space Exploration and Control for GANs

2025-09-26 · Kirsten Odendaal, Neela Kaushik, Spencer Halverson arxiv

This work integrates StyleGAN, DragGAN and Principal Component Analysis (PCA) to enhance the latent space efficiency and controllability of GAN-generated images. Style-GAN provides a structured latent space, DragGAN enab…

Dimensionality Reduction Image Manipulation