mlsanityData to 7 Oct 2026

Freeze the Discriminator: a Simple Baseline for Fine-Tuning GANs

arXiv 2002.10964 ↗First posted 25 Feb 2020cs.CV cs.LG stat.MLcited by 132

Not on the map. Why?

Abstract (arXiv)

Generative adversarial networks (GANs) have shown outstanding performance on a wide range of problems in computer vision, graphics, and machine learning, but often require numerous training data and heavy computational resources. To tackle this issue, several methods introduce a transfer learning technique in GAN training. They, however, are either prone to overfitting or limited to learning small distribution shifts. In this paper, we show that simple fine-tuning of GANs with frozen lower layers of the discriminator performs surprisingly well. This simple baseline, FreezeD, significantly outperforms previous techniques used in both unconditional and conditional GANs. We demonstrate the consistent effect using StyleGAN and SNGAN-projection architectures on several datasets of Animal Face, Anime Face, Oxford Flower, CUB-200-2011, and Caltech-256 datasets. The code and results are available at https://github.com/sangwoomo/FreezeD.

Key facts. Cited by 132 papers in the map (data through 7 Oct 2026). First posted on arXiv 25 Feb 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 26; 2022: 48; 2023: 35; 2024: 14; 2025: 6; 2026: 3.

132 papers.

Sort:

By area:

  1. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,286
  2. 2022Multi-Concept Customization of Text-to-Image Diffusioncited by 1,099
  3. 2023SVDiff: Compact Parameter Space for Diffusion Fine-Tuningcited by 338
  4. 2023Ablating Concepts in Text-to-Image Diffusion Modelscited by 312
  5. 2021Few-shot Image Generation via Cross-domain Correspondencecited by 206
  6. 2022Surgical Fine-Tuning Improves Adaptation to Distribution Shiftscited by 205
  7. 2021Projected GANs Converge Fastercited by 204
  8. 2021StyleGAN-NADA: CLIP-Guided Domain Adaptation of Image Generatorscited by 164
  9. 2021Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesiscited by 140
  10. 2023AnomalyDiffusion: Few-Shot Anomaly Image Generation with Diffusion Modelcited by 110
  11. 2021Regularizing Generative Adversarial Networks under Limited Datacited by 105
  12. 2021Ensembling Off-the-shelf Models for GAN Trainingcited by 84
38 more
  1. 2022Visual Prompt Tuning for Generative Transfer Learningcited by 75
  2. 2021Mind the Gap: Domain Gap Control for Single Shot Domain Adaptation for Generative Adversarial Networkscited by 68
  3. 2021GAN-Supervised Dense Visual Alignmentcited by 61
  4. 2021Rebooting ACGAN: Auxiliary Classifier GANs with Stable Trainingcited by 57
  5. 2022Few Shot Generative Model Adaption via Relaxed Spatial Structural Alignmentcited by 54
  6. 2021JoJoGAN: One Shot Face Stylizationcited by 49
  7. 2022A Closer Look at Few-shot Image Generationcited by 49
  8. 2021StyleAlign: Analysis and Applications of Aligned StyleGAN Modelscited by 48
  9. 20233D generation on ImageNetcited by 47
  10. 2021One-Shot Generative Domain Adaptationcited by 41
  11. 2022StudioGAN: A Taxonomy and Benchmark of GANs for Image Synthesiscited by 38
  12. 20233DAvatarGAN: Bridging Domains for Personalized Editable Avatarscited by 38
  13. 2021Sketch Your Own GANcited by 37
  14. 2022Diffusion Guided Domain Adaptation of Image Generatorscited by 35
  15. 2022Few-shot Image Generation via Adaptation-Aware Kernel Modulationcited by 33
  16. 2022DATID-3D: Diversity-Preserved Domain Adaptation Using Text-to-Image Diffusion for 3D Generative Modelcited by 32
  17. 2022One-Shot Adaptation of GAN in Just One CLIPcited by 31
  18. 2024Dual-Interrelated Diffusion Model for Few-Shot Anomaly Image Generationcited by 31
  19. 2022Fair Generative Models via Transfer Learningcited by 27
  20. 2023Zero-shot Generative Model Adaptation via Image-specific Prompt Learningcited by 27
  21. 2022MyStyle: A Personalized Generative Priorcited by 26
  22. 2021Data-Efficient GAN Training Beyond (Just) Augmentations: A Lottery Ticket Perspectivecited by 25
  23. 2022Rewriting Geometric Rules of a GANcited by 25
  24. 2022Collapse by Conditioning: Training Class-conditional GANs with Limited Datacited by 24
  25. 2022Towards Diverse and Faithful One-shot Adaption of Generative Adversarial Networkscited by 23
  26. 20223DFaceShop: Explicitly Controllable 3D-Aware Portrait Generationcited by 22
  27. 2022Few-shot Image Generation with Diffusion Modelscited by 21
  28. 2022A Comprehensive Survey on Data-Efficient GANs in Image Generationcited by 20
  29. 2022Improving GANs with A Dynamic Discriminatorcited by 20
  30. 2022Generalized One-shot Domain Adaptation of Generative Adversarial Networkscited by 19
  31. 2022HyperDomainNet: Universal Domain Adaptation for Generative Adversarial Networkscited by 19
  32. 2023Exploring Incompatible Knowledge Transfer in Few-shot Image Generationcited by 19
  33. 2023A Survey on Generative Modeling with Limited Data, Few Shots, and Zero Shotcited by 19
  34. 2022Data Redaction from Pre-trained GANscited by 18
  35. 2023Domain Re-Modulation for Few-Shot Generative Domain Adaptationcited by 18
  36. 2022CoordGAN: Self-Supervised Dense Correspondences Emerge from GANscited by 17
  37. 2022FakeCLR: Exploring Contrastive Learning for Solving Latent Discontinuity in Data-Efficient GANscited by 17
  38. 2021Few-shot Image Generation with Mixup-based Distance Learningcited by 16

50 most cited of 132 All 132 (JSON)