Object-Centric Representation Learning with Generative Spatial-Temporal Factorization

no code implementations • NeurIPS 2021 • Li Nanbo, Muhammad Ahmed Raza, Hu Wenbin, Zhaole Sun, Robert B. Fisher

We train DyMON on multi-view-dynamic-scene data and show that DyMON learns -- without supervision -- to factorize the entangled effects of observer motions and scene object dynamics from a sequence of observations, and constructs scene object spatial representations suitable for rendering at arbitrary times (querying across time) and from arbitrary viewpoints (querying across space).

Object Representation Learning