# Controllable Generative Adversarial Network

arXiv: https://arxiv.org/abs/1708.00598 · HTML: https://mlsanity.com/paper/1708.00598/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Aug 2017.

## Built on by (most cited first, 8 of 8)

- [MUSES: 3D-Controllable Image Generation via Multi-Modal Agent Collaboration](https://mlsanity.com/paper/2408.10605/) (2024)
- [MTVCraft: Tokenizing 4D Motion for Arbitrary Character Animation](https://mlsanity.com/paper/2505.10238/) (2025)
- [Enhanced Magnetic Resonance Image Synthesis with Contrast-Aware Generative Adversarial Networks](https://mlsanity.com/paper/2102.09386/) (2021)
- [LiteUpdate: A Lightweight Framework for Updating AI-Generated Image Detectors](https://mlsanity.com/paper/2511.07192/) (2025)
- [Adversarial Attention for Human Motion Synthesis](https://mlsanity.com/paper/2204.11751/) (2022)
- [DuDGAN: Improving Class-Conditional GANs via Dual-Diffusion](https://mlsanity.com/paper/2305.14849/) (2023)
- [Feature Space Topology Control via Hopkins Loss](https://mlsanity.com/paper/2509.11154/) (2025)
- [3D Underwater Path Planning via Generative Flow Field Surrogates](https://mlsanity.com/paper/2606.06077/) (2026)

## Abstract (arXiv)

Recently introduced generative adversarial network (GAN) has been shown numerous promising results to generate realistic samples. The essential task of GAN is to control the features of samples generated from a random distribution. While the current GAN structures, such as conditional GAN, successfully generate samples with desired major features, they often fail to produce detailed features that bring specific differences among samples. To overcome this limitation, here we propose a controllable GAN (ControlGAN) structure. By separating a feature classifier from a discriminator, the generator of ControlGAN is designed to learn generating synthetic samples with the specific detailed features. Evaluated with multiple image datasets, ControlGAN shows a power to generate improved samples with well-controlled features. Furthermore, we demonstrate that ControlGAN can generate intermediate features and opposite features for interpolated and extrapolated input labels that are not used in the training process. It implies that ControlGAN can significantly contribute to the variety of generated samples.