mlsanityData to 7 Oct 2026

Exploring Incompatible Knowledge Transfer in Few-shot Image Generation

arXiv 2304.07574 ↗First posted 15 Apr 2023cs.CVcited by 19

Not on the map. Why?

Abstract (arXiv)

Few-shot image generation (FSIG) learns to generate diverse and high-fidelity images from a target domain using a few (e.g., 10) reference samples. Existing FSIG methods select, preserve and transfer prior knowledge from a source generator (pretrained on a related domain) to learn the target generator. In this work, we investigate an underexplored issue in FSIG, dubbed as incompatible knowledge transfer, which would significantly degrade the realisticness of synthetic samples. Empirical observations show that the issue stems from the least significant filters from the source generator. To this end, we propose knowledge truncation to mitigate this issue in FSIG, which is a complementary operation to knowledge preservation and is implemented by a lightweight pruning-based method. Extensive experiments show that knowledge truncation is simple and effective, consistently achieving state-of-the-art performance, including challenging setups where the source and target domains are more distant. Project Page: yunqing-me.github.io/RICK.

Key facts. Cited by 19 papers in the map (data through 7 Oct 2026). Builds on 69 of its 96 references that are in the map. First posted on arXiv 15 Apr 2023.

Builds on

69 of 96 references.

References · 69

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  4. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  5. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  6. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  7. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  8. 2022Generative Adversarial Networkscited by 5,603
  9. 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
  10. 2017Neural Discrete Representation Learningcited by 4,650
  11. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
  12. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
38 more
  1. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  2. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
  3. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  4. 2015Variational Inference with Normalizing Flowscited by 2,161
  5. 2015Learning both Weights and Connections for Efficient Neural Networkscited by 1,998
  6. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,350
  7. 2018Demystifying MMD GANscited by 1,268
  8. 2014How transferable are features in deep neural networks?cited by 1,216
  9. 2020Training Generative Adversarial Networks with Limited Datacited by 1,184
  10. 2021Alias-Free Generative Adversarial Networkscited by 1,181
  11. 2019A Comprehensive Survey on Transfer Learningcited by 1,154
  12. 2019StarGAN v2: Diverse Image Synthesis for Multiple Domainscited by 1,095
  13. 2022Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distributioncited by 722
  14. 2017Network Dissection: Quantifying Interpretability of Deep Visual Representationscited by 703
  15. 2021Sparsity in Deep Learning: Pruning and growth for efficient inference and training in neural networkscited by 521
  16. 2019Quantifying the Carbon Emissions of Machine Learningcited by 516
  17. 2020In-Domain GAN Inversion for Real Image Editingcited by 420
  18. 2018Filter Pruning via Geometric Median for Deep Convolutional Neural Networks Accelerationcited by 365
  19. 2018GAN-based Synthetic Medical Image Augmentation for increased CNN Performance in Liver Lesion Classificationcited by 242
  20. 2018Explicit Inductive Bias for Transfer Learning with Convolutional Networkscited by 226
  21. 2013OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networkscited by 224
  22. 2021Few-shot Image Generation via Cross-domain Correspondencecited by 206
  23. 2019Task2Vec: Task Embedding for Meta-Learningcited by 197
  24. 2021High-Fidelity GAN Inversion for Image Attribute Editingcited by 197
  25. 2023A Recipe for Watermarking Diffusion Modelscited by 136
  26. 2018Transferring GANs: generating images from limited datacited by 134
  27. 2020Freeze the Discriminator: a Simple Baseline for Fine-Tuning GANscited by 132
  28. 2021OW-DETR: Open-world Detection Transformercited by 129
  29. 2022Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Differencecited by 123
  30. 2020Few-shot Image Generation with Elastic Weight Consolidationcited by 122
  31. 2019MineGAN: effective knowledge transfer from GANs to target domains with few imagescited by 106
  32. 2021Regularizing Generative Adversarial Networks under Limited Datacited by 105
  33. 2019Image Generation From Small Datasets via Batch Statistics Adaptationcited by 102
  34. 2022DiffPose: Toward More Reliable 3D Pose Estimationcited by 99
  35. 2020APQ: Joint Search for Network Architecture, Pruning and Quantization Policycited by 98
  36. 2021UAV-Human: A Large Benchmark for Human Behavior Understanding with Unmanned Aerial Vehiclescited by 95
  37. 2020GAN Memory with No Forgettingcited by 84
  38. 2022Head2Toe: Utilizing Intermediate Representations for Better Transfer Learningcited by 83

50 most cited of 69 All 69 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 6; 2024: 8; 2025: 4; 2026: 1.

19 papers.

Sort:

By area:

  1. 2023Label-Only Model Inversion Attacks via Knowledge Transfercited by 21
  2. 2023A Survey on Generative Modeling with Limited Data, Few Shots, and Zero Shotcited by 19
  3. 2024Deformable One-shot Face Stylization via DINO Semantic Guidancecited by 14
  4. 2023Image Synthesis under Limited Data: A Survey and Taxonomycited by 9
  5. 2024Generative Artificial Intelligence Meets Synthetic Aperture Radar: A Surveycited by 9
  6. 2023On Measuring Fairness in Generative Modelscited by 8
  7. 2024Few-Shot Image Generation by Conditional Relaxing Diffusion Inversioncited by 7
  8. 2025Few-shot Implicit Function Generation via Equivariancecited by 2
  9. 2025Uni-DAD: Unified Distillation and Adaptation of Diffusion Models for Few-step Few-shot Image Generationcited by 2
  10. 2023Few-shot 3D Shape Generationcited by 1
  11. 2023AdAM: Few-Shot Image Generation via Adaptation-Aware Kernel Modulationcited by 1
  12. 2024A Universal Knowledge Embedded Contrastive Learning Framework for Hyperspectral Image Classificationcited by 1
7 more
  1. 2024DomainGallery: Few-shot Domain-driven Image Generation by Attribute-centric Finetuningcited by 1
  2. 2024UniHDA: A Unified and Versatile Framework for Multi-Modal Hybrid Domain Adaptationcited by 0
  3. 2024On The Relationship between Visual Anomaly-free and Anomalous Representationscited by 0
  4. 2024TAGE: Trustworthy Attribute Group Editing for Stable Few-shot Image Generationcited by 0
  5. 2025AIR: Zero-shot Generative Model Adaptation with Iterative Refinementcited by 0
  6. 2025A Turn Toward Better Alignment: Few-Shot Generative Adaptation with Equivariant Feature Rotationcited by 0
  7. 2026Few-Shot Generative Model Adaption via Identity Injection and Preservationcited by 0