mlsanityData to 14 Oct 2026

Sequential Representation Learning via Static-Dynamic Conditional Disentanglement

arXiv 2408.05599 ↗First posted 10 Aug 2024cs.LG cs.AI cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

This paper explores self-supervised disentangled representation learning within sequential data, focusing on separating time-independent and time-varying factors in videos. We propose a new model that breaks the usual independence assumption between those factors by explicitly accounting for the causal relationship between the static/dynamic variables and that improves the model expressivity through additional Normalizing Flows. A formal definition of the factors is proposed. This formalism leads to the derivation of sufficient conditions for the ground truth factors to be identifiable, and to the introduction of a novel theoretically grounded disentanglement constraint that can be directly and efficiently incorporated into our new framework. The experiments show that the proposed approach outperforms previous complex state-of-the-art techniques in scenarios where the dynamics of a scene are influenced by its content.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 44 of its 58 references that are in the map. First posted on arXiv 10 Aug 2024.

Builds on

44 of 58 references.

References · 44

  1. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  2. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,588
  3. 2015Variational Inference with Normalizing Flowscited by 2,170
  4. 2014NICE: Non-linear Independent Components Estimationcited by 1,058
  5. 2018Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representationscited by 941
  6. 2016Improving Variational Inference with Inverse Autoregressive Flowcited by 640
  7. 2020NVAE: A Deep Hierarchical Variational Autoencodercited by 637
  8. 2018Disentangling by Factorisingcited by 616
  9. 2017MoCoGAN: Decomposing Motion and Content for Video Generationcited by 524
  10. 2018Neural Processescited by 311
  11. 2020Weakly-Supervised Disentanglement Without Compromisescited by 239
  12. 2018Neural Autoregressive Flowscited by 208
32 more
  1. 2016Variational Lossy Autoencodercited by 176
  2. 2019Learning Likelihoods with Conditional Normalizing Flowscited by 172
  3. 2017Decomposing Motion and Content for Natural Video Sequence Predictioncited by 169
  4. 2021A Good Image Generator Is What You Need for High-Resolution Video Synthesiscited by 163
  5. 2022Weakly supervised causal representation learningcited by 140
  6. 2017Unsupervised Learning of Disentangled Representations from Videocited by 139
  7. 2022CITRIS: Causal Identifiability from Temporal Intervened Sequencescited by 133
  8. 2018Disentangled Sequential Autoencodercited by 120
  9. 2022Disentangled Representation Learningcited by 112
  10. 2017Multi-Level Variational Autoencoder: Learning Disentangled Representations from Grouped Observationscited by 100
  11. 2019On the Transfer of Inductive Bias from Simulation to the Real World: a New Disentanglement Datasetcited by 100
  12. 2017Unsupervised Learning of Disentangled and Interpretable Representations from Sequential Datacited by 93
  13. 2020S3VAE: Self-Supervised Sequential VAE for Representation Disentanglement and Data Generationcited by 63
  14. 2019Disentangling Factors of Variation Using Few Labelscited by 57
  15. 2020CausalVAE: Structured Causal Disentanglement in Variational Autoencodercited by 48
  16. 2021Learning Disentangled Representations in the Imaging Domaincited by 46
  17. 2021Contrastively Disentangled Sequential Variational Autoencodercited by 34
  18. 2019Demystifying Inter-Class Disentanglementcited by 31
  19. 2021Commutative Lie Group VAE for Disentanglement Learningcited by 27
  20. 2023Multifactor Sequential Disentanglement via Structured Koopman Autoencoderscited by 24
  21. 2021Disentangled Recurrent Wasserstein Autoencodercited by 22
  22. 2022Decoupling Local and Global Representations of Time Seriescited by 17
  23. 2023Sample and Predict Your Latent: Modality-free Sequential Disentanglement via Contrastive Estimationcited by 10
  24. 2022Review of Disentanglement Approaches for Medical Applications -- Towards Solving the Gordian Knot of Generative Models in Healthcarecited by 9
  25. 2020Improving Sequential Latent Variable Models with Autoregressive Flowscited by 8
  26. 2020Variational Autoencoders with Normalizing Flow Decoderscited by 7
  27. 2022A unified model for continuous conditional video predictioncited by 7
  28. 2022Retriever: Learning Content-Style Representation as a Token-Level Bipartite Graphcited by 6
  29. 2022Towards Robust Unsupervised Disentanglement of Sequential Data -- A Case Study Using Music Audiocited by 6
  30. 2018Scalable Factorized Hierarchical Variational Autoencoder Trainingcited by 5
  31. 2021Video Reenactment as Inductive Bias for Content-Motion Disentanglementcited by 3
  32. 2016Deep Variational Inference Without Pixel-Wise Reconstructioncited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 1.

4 papers.

  1. 2025Dyn-O: Building Structured World Models with Object-Centric Representationscited by 13
  2. 2025Bitrate-Controlled Diffusion for Disentangling Motion and Content in Videocited by 2
  3. 2025Disentanglement Beyond Static vs. Dynamic: A Benchmark and Evaluation Framework for Multi-Factor Sequential Representationscited by 1
  4. 2026Unsupervised Disentanglement Without Compromises : How Functional Orthogonality Enforces Identifiabilitycited by 0