arXiv · 2006.14727
Unsupervised Video Decomposition using Spatio-temporal Iterative Inference
Abstract
Unsupervised multi-object scene decomposition is a fast-emerging problem in representation learning. Despite significant progress in static scenes, such models are unable to leverage important dynamic cues present in video. We propose a novel spatio-temporal iterative inference framework that is powerful enough to jointly model complex multi-object representations and explicit temporal dependencies between latent variables across frames. This is achieved by leveraging 2D-LSTM, temporally conditioned inference and generation within the iterative amortized inference for posterior refinement. Our method improves the overall quality of decompositions, encodes information about the objects' dynamics, and can be used to predict trajectories of each object separately. Additionally, we show that our model has a high accuracy even without color information. We demonstrate the decomposition, segmentation, and prediction capabilities of our model and show that it outperforms the state-of-the-art on several benchmark datasets, one of which was curated for this work and will be made publicly available.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Polina Zablotskaia, Edoardo A. Dominici, Leonid Sigal, Andreas M. Lehrmann. 2020-06-25. Unsupervised Video Decomposition using Spatio-temporal Iterative Inference. https://arxiv.org/abs/2006.14727
Cite the original work for its findings. Save a collection to share your selection of sources.