mlsanityData to 7 Oct 2026

Unsupervised Disentanglement of Linear-Encoded Facial Semantics

arXiv 2103.16605 ↗First posted 30 Mar 2021cs.CVcited by 7

Not on the map. Why?

Abstract (arXiv)

We propose a method to disentangle linear-encoded facial semantics from StyleGAN without external supervision. The method derives from linear regression and sparse representation learning concepts to make the disentangled latent representations easily interpreted as well. We start by coupling StyleGAN with a stabilized 3D deformable facial reconstruction method to decompose single-view GAN generations into multiple semantics. Latent representations are then extracted to capture interpretable facial semantics. In this work, we make it possible to get rid of labels for disentangling meaningful facial semantics. Also, we demonstrate that the guided extrapolation along the disentangled representations can help with data augmentation, which sheds light on handling unbalanced data. Finally, we provide an analysis of our learned localized facial representations and illustrate that the semantic information is encoded, which surprisingly complies with human intuition. The overall unsupervised design brings more flexibility to representation learning in the wild.

Key facts. Cited by 7 papers in the map (data through 7 Oct 2026). Builds on 32 of its 45 references that are in the map. First posted on arXiv 30 Mar 2021.

Builds on

32 of 45 references.

References · 32

  1. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  2. 2017Neural Discrete Representation Learningcited by 4,650
  3. 2013Auto-Encoding Variational Bayescited by 3,980
  4. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  5. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
  6. 2017Improved Training of Wasserstein GANscited by 2,223
  7. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,618
  8. 2017StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translationcited by 1,036
  9. 2018Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representationscited by 929
  10. 2016InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Netscited by 858
  11. 2018Self-Attention Generative Adversarial Networkscited by 809
  12. 2019Interpreting the Latent Space of GANs for Semantic Face Editingcited by 691
20 more
  1. 2020NVAE: A Deep Hierarchical Variational Autoencodercited by 635
  2. 2018Disentangling by Factorisingcited by 614
  3. 2016Conditional Image Synthesis With Auxiliary Classifier GANscited by 562
  4. 2017How far are we from solving the 2D & 3D Face Alignment problem? (and a dataset of 230,000 3D facial landmarks)cited by 524
  5. 2014Learning Face Representation from Scratchcited by 451
  6. 2020InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANscited by 393
  7. 2019On the "steerability" of generative adversarial networkscited by 247
  8. 2020Disentangled and Controllable Face Image Generation via 3D Imitative-Contrastive Learningcited by 219
  9. 2019Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wildcited by 180
  10. 2018Isolating Sources of Disentanglement in Variational Autoencoderscited by 167
  11. 2017Optimizing the Latent Space of Generative Networkscited by 157
  12. 2019STGAN: A Unified Selective Transfer Network for Arbitrary Image Attribute Editingcited by 119
  13. 2017Fader Networks: Manipulating Images by Sliding Attributescited by 118
  14. 2019Semantic Hierarchy Emerges in Deep Generative Representations for Scene Synthesiscited by 106
  15. 2017The Riemannian Geometry of Deep Generative Modelscited by 81
  16. 2018Towards Open-Set Identity Preserving Face Synthesiscited by 74
  17. 2017Towards Large-Pose Face Frontalization in the Wildcited by 50
  18. 2017Semantically Decomposing the Latent Spaces of Generative Adversarial Networkscited by 19
  19. 2019InfoGAN-CR and ModelCentrality: Self-supervised Model Training and Selection for Disentangling GANscited by 16
  20. 2018FaceFeat-GAN: a Two-Stage Approach for Identity-Preserving Face Synthesiscited by 14

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 2; 2023: 0; 2024: 2; 2025: 2.

7 papers.

  1. 2022Progressive Disentangled Representation Learning for Fine-Grained Controllable Talking Head Synthesiscited by 93
  2. 2022Extracting Semantic Knowledge from GANs with Unsupervised Learningcited by 5
  3. 2021Semantic and Geometric Unfolding of StyleGAN Latent Spacecited by 3
  4. 2025CAMS: Towards Compositional Zero-Shot Learning via Gated Cross-Attention and Multi-Space Disentanglementcited by 1
  5. 2024DrFER: Learning Disentangled Representations for 3D Facial Expression Recognitioncited by 0
  6. 2024A Reference-Based 3D Semantic-Aware Framework for Accurate Local Facial Attribute Editingcited by 0
  7. 2025Bringing Diversity from Diffusion Models to Semantic-Guided Face Asset Generationcited by 0