Unsupervised multi-object scene decomposition is a fast-emerging problem in representation learning. Despite significant progress in static scenes, models are unable to leverage important dynamic cues present in video.
BibTex:
Before browse our site, please accept our cookies policy
