mlsanityData to 7 Oct 2026

Transferring disentangled representations: bridging the gap between synthetic and real images

arXiv 2409.18017 ↗First posted 26 Sep 2024cs.CV cs.AI cs.LGcited by 1

Not on the map. Why?

Abstract (arXiv)

Developing meaningful and efficient representations that separate the fundamental structure of the data generation mechanism is crucial in representation learning. However, Disentangled Representation Learning has not fully shown its potential on real images, because of correlated generative factors, their resolution and limited access to ground truth labels. Specifically on the latter, we investigate the possibility of leveraging synthetic data to learn general-purpose disentangled representations applicable to real data, discussing the effect of fine-tuning and what properties of disentanglement are preserved after the transfer. We provide an extensive empirical study to address these issues. In addition, we propose a new interpretable intervention-based metric, to measure the quality of factors encoding in the representation. Our results indicate that some level of disentanglement, transferring a representation from synthetic to real data, is possible and effective.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 49 of its 72 references that are in the map. First posted on arXiv 26 Sep 2024.

Builds on

49 of 72 references.

References · 49

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2018Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representationscited by 929
  3. 2016InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Netscited by 858
  4. 2018Disentangling by Factorisingcited by 614
  5. 2015Generating Sentences from a Continuous Spacecited by 607
  6. 2018Understanding disentangling in β-VAEcited by 332
  7. 2016Ladder Variational Autoencoderscited by 315
  8. 2018Towards a Definition of Disentangled Representationscited by 294
  9. 2017Variational Inference of Disentangled Latent Concepts from Unlabeled Observationscited by 240
  10. 2020Weakly-Supervised Disentanglement Without Compromisescited by 237
  11. 2019On the Fairness of Disentangled Representationscited by 144
  12. 2018Visual Object Networks: Image Generation with Disentangled 3D Representationcited by 124
37 more
  1. 2015Deep Convolutional Inverse Graphics Networkcited by 121
  2. 2019Spatial Broadcast Decoder: A Simple Architecture for Learning Disentangled Representations in VAEscited by 116
  3. 2018Robustly Disentangled Causal Mechanisms: Validating Deep Representations for Interventional Robustnesscited by 108
  4. 2018Learning Deep Disentangled Embeddings with the F-Statistic Losscited by 101
  5. 2017Multi-Level Variational Autoencoder: Learning Disentangled Representations from Grouped Observationscited by 100
  6. 2019Are Disentangled Representations Helpful for Abstract Visual Reasoning?cited by 100
  7. 2019On the Transfer of Inductive Bias from Simulation to the Real World: a New Disentanglement Datasetcited by 100
  8. 2017Disentangled Person Image Generationcited by 98
  9. 2020On Disentangled Representations Learned From Correlated Datacited by 94
  10. 2019Weakly Supervised Disentanglement with Guaranteescited by 82
  11. 2020Fairness by Learning Orthogonal Disentangled Representationscited by 76
  12. 2020Measuring Disentanglement: A Review of Metricscited by 69
  13. 2020On the Transfer of Disentangled Representations in Realistic Settingscited by 63
  14. 2018Image-to-image translation for cross-domain disentanglementcited by 61
  15. 2021SemanticStyleGAN: Learning Compositional Generative Priors for Controllable Image Synthesis and Editingcited by 58
  16. 2019Theory and Evaluation Metrics for Learning Disentangled Representationscited by 57
  17. 2023Posterior Collapse and Latent Variable Non-identifiabilitycited by 57
  18. 2020Semi-Supervised StyleGAN for Disentanglement Learningcited by 43
  19. 2023Disentanglement via Latent Quantizationcited by 40
  20. 2021DRANet: Disentangling Representation and Adaptation Networks for Unsupervised Cross-Domain Adaptationcited by 37
  21. 2021Smoothing the Disentangled Latent Style Space for Unsupervised Image-to-Image Translationcited by 35
  22. 2023Leveraging sparse and shared feature activations for disentangled representation learningcited by 32
  23. 2023DisDiff: Unsupervised Disentanglement of Diffusion Probabilistic Modelscited by 30
  24. 2018Improving Generalization for Abstract Reasoning Tasks Using Disentangled Feature Representationscited by 28
  25. 2021Where and What? Examining Interpretable Disentangled Representationscited by 27
  26. 2016A Survey of Inductive Biases for Factorial Representation-Learningcited by 23
  27. 2019Relevance Factor VAE: Learning and Identifying Disentangled Factorscited by 20
  28. 2021Learning disentangled representations via product manifold projectioncited by 20
  29. 2023Hierarchical Diffusion Autoencoders and Disentangled Image Manipulationcited by 18
  30. 2020PCAAE: Principal Component Analysis Autoencoder for organising the latent space of generative networkscited by 16
  31. 2020Disentangled Representation Learning and Generation with Manifold Optimizationcited by 15
  32. 2021Scaling-up Disentanglement for Image Translationcited by 15
  33. 2022A Contrastive Objective for Learning Disentangled Representationscited by 14
  34. 2022An Empirical Study on Disentanglement of Negative-free Contrastive Learningcited by 12
  35. 2023Flow Factorized Representation Learningcited by 11
  36. 2021DisUnknown: Distilling Unknown Factors for Disentanglement Learningcited by 6
  37. 2023Beyond Vanilla Variational Autoencoders: Detecting Posterior Collapse in Conditional and Hierarchical Variational Autoencoderscited by 6

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025Disentangled representations of microscopy imagescited by 0