mlsanityData to 7 Oct 2026

Interpreting the Latent Space of GANs for Semantic Face Editing

arXiv 1907.10786 ↗First posted 25 Jul 2019cs.CVcited by 691

Not on the map. Why?

Cited by 95 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Despite the recent advance of Generative Adversarial Networks (GANs) in high-fidelity image synthesis, there lacks enough understanding of how GANs are able to map a latent code sampled from a random distribution to a photo-realistic image. Previous work assumes the latent space learned by GANs follows a distributed representation but observes the vector arithmetic phenomenon. In this work, we propose a novel framework, called InterFaceGAN, for semantic face editing by interpreting the latent semantics learned by GANs. In this framework, we conduct a detailed study on how different semantics are encoded in the latent space of GANs for face synthesis. We find that the latent code of well-trained generative models actually learns a disentangled representation after linear transformations. We explore the disentanglement between various semantics and manage to decouple some entangled semantics with subspace projection, leading to more precise control of facial attributes. Besides manipulating gender, age, expression, and the presence of eyeglasses, we can even vary the face pose as well as fix the artifacts accidentally generated by GAN models. The proposed method is further applied to achieve real image manipulation when combined with GAN inversion methods or some encoder-involved models. Extensive results suggest that learning to synthesize faces spontaneously brings a disentangled and controllable facial attribute representation.

Key facts. Cited by 691 papers in the map (data through 7 Oct 2026). First posted on arXiv 25 Jul 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 153; 2022: 173; 2023: 166; 2024: 112; 2025: 60; 2026: 27.

691 papers.

Sort:

By area:

  1. 2021SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equationscited by 1,903
  2. 2022Imagic: Text-Based Real Image Editing with Diffusion Modelscited by 1,172
  3. 2022Diffusion Models in Vision: A Surveycited by 1,008
  4. 2021StyleCLIP: Text-Driven Manipulation of StyleGAN Imagerycited by 990
  5. 2023GLIGEN: Open-Set Grounded Text-to-Image Generationcited by 774
  6. 2022DiffEdit: Diffusion-based semantic image editing with mask guidancecited by 643
  7. 2021ILVR: Conditioning Method for Denoising Diffusion Probabilistic Modelscited by 634
  8. 2021Designing an Encoder for StyleGAN Image Manipulationcited by 542
  9. 2022Paint by Example: Exemplar-based Image Editing with Diffusion Modelscited by 489
  10. 2021StyleNeRF: A Style-based 3D-Aware Generator for High-resolution Image Synthesiscited by 477
  11. 2021Diffusion Autoencoders: Toward a Meaningful and Decodable Representationcited by 447
  12. 2021Pivotal Tuning for Latent-based Editing of Real Imagescited by 427
38 more
  1. 2021GAN Inversion: A Surveycited by 366
  2. 2022From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AIcited by 299
  3. 2023DragDiffusion: Harnessing Diffusion Models for Interactive Point-based Image Editingcited by 293
  4. 2021StyleSDF: High-Resolution 3D-Consistent Image and Geometry Generationcited by 289
  5. 2021ReStyle: A Residual-Based StyleGAN Encoder via Iterative Refinementcited by 248
  6. 2023Drag Your GAN: Interactive Point-based Manipulation on the Generative Image Manifoldcited by 218
  7. 2023StyleGAN-T: Unlocking the Power of GANs for Fast Large-Scale Text-to-Image Synthesiscited by 212
  8. 2024Diffusion Model-Based Image Editing: A Surveycited by 212
  9. 2021HyperStyle: StyleGAN Inversion with HyperNetworks for Real Image Editingcited by 208
  10. 2023Encoder-based Domain Tuning for Fast Personalization of Text-to-Image Modelscited by 203
  11. 2021High-Fidelity GAN Inversion for Image Attribute Editingcited by 197
  12. 2022Latent Image Animator: Learning to Animate Images via Latent Space Navigationcited by 172
  13. 2021StyleGAN-NADA: CLIP-Guided Domain Adaptation of Image Generatorscited by 164
  14. 2021A Good Image Generator Is What You Need for High-Resolution Video Synthesiscited by 162
  15. 2021EditGAN: High-Precision Semantic Image Editingcited by 156
  16. 2021CIPS-3D: A 3D-Aware Generator of GANs Based on Conditionally-Independent Pixel Synthesiscited by 149
  17. 2023Concept Sliders: LoRA Adaptors for Precise Control in Diffusion Modelscited by 148
  18. 2023Localizing Object-level Shape Variations with Text-to-Image Diffusion Modelscited by 142
  19. 2021Frequency-aware Discriminative Feature Learning Supervised by Single-Center Loss for Face Forgery Detectioncited by 133
  20. 2021PSCC-Net: Progressive Spatio-Channel Correlation Network for Image Manipulation Detection and Localizationcited by 133
  21. 2021Explaining in Style: Training a GAN to explain a classifier in StyleSpacecited by 122
  22. 2022Uncovering the Disentanglement Capability in Text-to-Image Diffusion Modelscited by 116
  23. 2023Collaborative Diffusion for Multi-Modal Face Generation and Editingcited by 116
  24. 2023Vox-E: Text-guided Voxel Editing of 3D Objectscited by 114
  25. 2021A-SDF: Learning Disentangled Signed Distance Functions for Articulated Shape Representationcited by 113
  26. 2021Paint by Wordcited by 107
  27. 2021One Shot Face Swapping on Megapixelscited by 105
  28. 2021Generative Models as a Data Source for Multiview Representation Learningcited by 104
  29. 2021Deep Generative Learning via Schr"odinger Bridgecited by 103
  30. 2021Talk-to-Edit: Fine-Grained Facial Editing via Dialogcited by 102
  31. 2021D2C: Diffusion-Denoising Models for Few-shot Conditional Generationcited by 96
  32. 2022Progressive Disentangled Representation Learning for Fine-Grained Controllable Talking Head Synthesiscited by 93
  33. 2021Only a Matter of Style: Age Transformation Using a Style-Based Regression Modelcited by 92
  34. 2021GAN-Control: Explicitly Controllable GANscited by 91
  35. 2021Exploiting Spatial Dimensions of Latent in GAN for Real-time Image Editingcited by 91
  36. 2021Towards Principled Disentanglement for Domain Generalizationcited by 91
  37. 2022Language Models Can See: Plugging Visual Controls in Text Generationcited by 89
  38. 2021Repurposing GANs for One-shot Semantic Part Segmentationcited by 88

50 most cited of 691 All 691 (JSON)