Sequential Representation Learning via Static-Dynamic Conditional Disentanglement
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.
Code (0)
등록된 구현이 없습니다.
Tasks
DisentanglementRepresentation LearningMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
FAVAE: SEQUENCE DISENTANGLEMENT USING IN- FORMATION BOTTLENECK PRINCIPLE
A state-of-the-art generative model, a ”factorized action variational autoencoder (FAVAE),” is presented for learning disentangled and interpretable representations from sequential data via the information bottleneck wit…
DisentanglementRepresentation LearningDisentangled Recurrent Wasserstein Autoencoder
Learning disentangled representations leads to interpretable models and facilitates data generation with style transfer, which has been extensively studied on static data such as images in an unsupervised learning framew…
DisentanglementRepresentation LearningStyle TransferUnconditional Video Generation+1FAVAE: Sequence Disentanglement using Information Bottleneck Principle
We propose the factorized action variational autoencoder (FAVAE), a state-of-the-art generative model for learning disentangled and interpretable representations from sequential data via the information bottleneck withou…
DisentanglementRepresentation LearningDiViD: Disentangled Video Diffusion for Static-Dynamic Factorization
Unsupervised disentanglement of static appearance and dynamic motion in video remains a fundamental challenge, often hindered by information leakage and blurry reconstructions in existing VAE- and GAN-based approaches. W…
Disentanglement Beyond Static vs. Dynamic: A Benchmark and Evaluation Framework for Multi-Factor Sequential Representations
Learning disentangled representations in sequential data is a key goal in deep learning, with broad applications in vision, audio, and time series. While real-world data involves multiple interacting semantic factors ove…