# Face Identity Disentanglement via Latent Space Mapping

arXiv: https://arxiv.org/abs/2005.07728 · HTML: https://mlsanity.com/paper/2005.07728/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 May 2020.

## Built on by (most cited first, 10 of 10)

- [GAN Inversion: A Survey](https://mlsanity.com/paper/2101.05278/) (2021)
- [ReStyle: A Residual-Based StyleGAN Encoder via Iterative Refinement](https://mlsanity.com/paper/2104.02699/) (2021)
- [One Shot Face Swapping on Megapixels](https://mlsanity.com/paper/2105.04932/) (2021)
- [Picture that Sketch: Photorealistic Image Generation from Abstract Sketches](https://mlsanity.com/paper/2303.11162/) (2023)
- [Face Identity-Aware Disentanglement in StyleGAN](https://mlsanity.com/paper/2309.12033/) (2023)
- [FaceDNeRF: Semantics-Driven Face Reconstruction, Prompt Editing and Relighting with Diffusion Models](https://mlsanity.com/paper/2306.00783/) (2023)
- [PluGeN: Multi-Label Conditional Generation From Pre-Trained Models](https://mlsanity.com/paper/2109.09011/) (2021)
- [Perceptually Validated Precise Local Editing for Facial Action Units with StyleGAN](https://mlsanity.com/paper/2107.12143/) (2021)
- [Face Cartoonisation For Various Poses Using StyleGAN](https://mlsanity.com/paper/2309.14908/) (2023)
- [SwinStyleformer is a favorable choice for image inversion](https://mlsanity.com/paper/2406.13153/) (2024)

## Abstract (arXiv)

Learning disentangled representations of data is a fundamental problem in artificial intelligence. Specifically, disentangled latent representations allow generative models to control and compose the disentangled factors in the synthesis process. Current methods, however, require extensive supervision and training, or instead, noticeably compromise quality. In this paper, we present a method that learns how to represent data in a disentangled way, with minimal supervision, manifested solely using available pre-trained networks. Our key insight is to decouple the processes of disentanglement and synthesis, by employing a leading pre-trained unconditional image generator, such as StyleGAN. By learning to map into its latent space, we leverage both its state-of-the-art quality, and its rich and expressive latent space, without the burden of training it. We demonstrate our approach on the complex and high dimensional domain of human heads. We evaluate our method qualitatively and quantitatively, and exhibit its success with de-identification operations and with temporal identity coherency in image sequences. Through extensive experimentation, we show that our method successfully disentangles identity from other facial attributes, surpassing existing methods, even though they require more training and supervision.