# GAN "Steerability" without optimization

arXiv: https://arxiv.org/abs/2012.05328 · HTML: https://mlsanity.com/paper/2012.05328/

## Key facts

- Cited by 48 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Dec 2020.

## Built on by (most cited first, 30 of 48)

- [Designing an Encoder for StyleGAN Image Manipulation](https://mlsanity.com/paper/2102.02766/) (2021)
- [StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasets](https://mlsanity.com/paper/2202.00273/) (2022)
- [Pivotal Tuning for Latent-based Editing of Real Images](https://mlsanity.com/paper/2106.05744/) (2021)
- [GAN Inversion: A Survey](https://mlsanity.com/paper/2101.05278/) (2021)
- [HyperStyle: StyleGAN Inversion with HyperNetworks for Real Image Editing](https://mlsanity.com/paper/2111.15666/) (2021)
- [GAN-Control: Explicitly Controllable GANs](https://mlsanity.com/paper/2101.02477/) (2021)
- [Discovering Interpretable Directions in the Semantic Latent Space of Diffusion Models](https://mlsanity.com/paper/2303.11073/) (2023)
- [Low-Rank Subspaces in GANs](https://mlsanity.com/paper/2106.04488/) (2021)
- [LatentCLR: A Contrastive Learning Approach for Unsupervised Discovery of Interpretable Directions](https://mlsanity.com/paper/2104.00820/) (2021)
- [WarpedGANSpace: Finding non-linear RBF paths in GAN latent space](https://mlsanity.com/paper/2109.13357/) (2021)
- [TransEditor: Transformer-Based Dual-Space GAN for Highly Controllable Facial Editing](https://mlsanity.com/paper/2203.17266/) (2022)
- [State-of-the-Art in the Architecture, Methods and Applications of StyleGAN](https://mlsanity.com/paper/2202.14020/) (2022)
- [User-Controllable Latent Transformer for StyleGAN Image Layout Editing](https://mlsanity.com/paper/2208.12408/) (2022)
- [MyStyle: A Personalized Generative Prior](https://mlsanity.com/paper/2203.17272/) (2022)
- [Region-Based Semantic Factorization in GANs](https://mlsanity.com/paper/2202.09649/) (2022)
- [The Euclidean Space is Evil: Hyperbolic Attribute Editing for Few-shot Image Generation](https://mlsanity.com/paper/2211.12347/) (2022)
- [Everything is There in Latent Space: Attribute Editing and Attribute Style Manipulation by StyleGAN Latent Space Exploration](https://mlsanity.com/paper/2207.09855/) (2022)
- [LARGE: Latent-Based Regression through GAN Semantics](https://mlsanity.com/paper/2107.11186/) (2021)
- [Latent Traversals in Generative Models as Potential Flows](https://mlsanity.com/paper/2304.12944/) (2023)
- [LinkGAN: Linking GAN Latents to Pixels for Controllable Image Synthesis](https://mlsanity.com/paper/2301.04604/) (2023)
- [Attribute-specific Control Units in StyleGAN for Fine-grained Image Manipulation](https://mlsanity.com/paper/2111.13010/) (2021)
- [Domain Expansion of Image Generators](https://mlsanity.com/paper/2301.05225/) (2023)
- [Latent Transformations via NeuralODEs for GAN-based Image Editing](https://mlsanity.com/paper/2111.14825/) (2021)
- [Masked-Attention Diffusion Guidance for Spatially Controlling Text-to-Image Generation](https://mlsanity.com/paper/2308.06027/) (2023)
- [IA-FaceS: A Bidirectional Method for Semantic Face Editing](https://mlsanity.com/paper/2203.13097/) (2022)
- [Learning Foreground-Background Segmentation from Improved Layered GANs](https://mlsanity.com/paper/2104.00483/) (2021)
- [Cycle Encoding of a StyleGAN Encoder for Improved Reconstruction and Editability](https://mlsanity.com/paper/2207.09367/) (2022)
- [What Decreases Editing Capability? Domain-Specific Hybrid Refinement for Improved GAN Inversion](https://mlsanity.com/paper/2301.12141/) (2023)
- [Householder Projector for Unsupervised Latent Semantics Discovery](https://mlsanity.com/paper/2307.08012/) (2023)
- [Multi-Attribute Balanced Sampling for Disentangled GAN Controls](https://mlsanity.com/paper/2111.00909/) (2021)

## Abstract (arXiv)

Recent research has shown remarkable success in revealing "steering" directions in the latent spaces of pre-trained GANs. These directions correspond to semantically meaningful image transformations e.g., shift, zoom, color manipulations), and have similar interpretable effects across all categories that the GAN can generate. Some methods focus on user-specified transformations, while others discover transformations in an unsupervised manner. However, all existing techniques rely on an optimization procedure to expose those directions, and offer no control over the degree of allowed interaction between different transformations. In this paper, we show that "steering" trajectories can be computed in closed form directly from the generator's weights without any form of training or optimization. This applies to user-prescribed geometric transformations, as well as to unsupervised discovery of more complex effects. Our approach allows determining both linear and nonlinear trajectories, and has many advantages over previous methods. In particular, we can control whether one transformation is allowed to come on the expense of another (e.g. zoom-in with or without allowing translation to keep the object centered). Moreover, we can determine the natural end-point of the trajectory, which corresponds to the largest extent to which a transformation can be applied without incurring degradation. Finally, we show how transferring attributes between images can be achieved without optimization, even across different categories.