# The Devil is in the Details: StyleFeatureEditor for Detail-Rich StyleGAN Inversion and High Quality Image Editing

arXiv: https://arxiv.org/abs/2406.10601 · HTML: https://mlsanity.com/paper/2406.10601/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 48 references that are in the map.
- First posted on arXiv 15 Jun 2024.

## Builds on (most cited first, 7 of 7)

- [On the "steerability" of generative adversarial networks](https://mlsanity.com/paper/1907.07171/) (2019)
- [Controlling generative models with continuous factors of variations](https://mlsanity.com/paper/2001.10238/) (2020)
- [GAN "Steerability" without optimization](https://mlsanity.com/paper/2012.05328/) (2020)
- [Feature-Style Encoder for Style-Based GAN Inversion](https://mlsanity.com/paper/2202.02183/) (2022)
- [Barbershop: GAN-based Image Compositing using Segmentation Masks](https://mlsanity.com/paper/2106.01505/) (2021)
- [Make It So: Steering StyleGAN for Any Image Inversion and Editing](https://mlsanity.com/paper/2304.14403/) (2023)
- [What Decreases Editing Capability? Domain-Specific Hybrid Refinement for Improved GAN Inversion](https://mlsanity.com/paper/2301.12141/) (2023)

## Abstract (arXiv)

The task of manipulating real image attributes through StyleGAN inversion has been extensively researched. This process involves searching latent variables from a well-trained StyleGAN generator that can synthesize a real image, modifying these latent variables, and then synthesizing an image with the desired edits. A balance must be struck between the quality of the reconstruction and the ability to edit. Earlier studies utilized the low-dimensional W-space for latent search, which facilitated effective editing but struggled with reconstructing intricate details. More recent research has turned to the high-dimensional feature space F, which successfully inverses the input image but loses much of the detail during editing. In this paper, we introduce StyleFeatureEditor -- a novel method that enables editing in both w-latents and F-latents. This technique not only allows for the reconstruction of finer image details but also ensures their preservation during editing. We also present a new training pipeline specifically designed to train our model to accurately edit F-latents. Our method is compared with state-of-the-art encoding approaches, demonstrating that our model excels in terms of reconstruction quality and is capable of editing even challenging out-of-domain examples. Code is available at https://github.com/AIRI-Institute/StyleFeatureEditor.