Timezone: »

Object Scene Representation Transformer
Mehdi S. M. Sajjadi · Daniel Duckworth · Aravindh Mahendran · Sjoerd van Steenkiste · Filip Pavetic · Mario Lucic · Leonidas Guibas · Klaus Greff · Thomas Kipf

Wed Nov 30 02:00 PM -- 04:00 PM (PST) @ Hall J #638

A compositional understanding of the world in terms of objects and their geometry in 3D space is considered a cornerstone of human cognition. Facilitating the learning of such a representation in neural networks holds promise for substantially improving labeled data efficiency. As a key step in this direction, we make progress on the problem of learning 3D-consistent decompositions of complex scenes into individual objects in an unsupervised fashion. We introduce Object Scene Representation Transformer (OSRT), a 3D-centric model in which individual object representations naturally emerge through novel view synthesis. OSRT scales to significantly more complex scenes with larger diversity of objects and backgrounds than existing methods. At the same time, it is multiple orders of magnitude faster at compositional rendering thanks to its light field parametrization and the novel Slot Mixer decoder. We believe this work will not only accelerate future architecture exploration and scaling efforts, but it will also serve as a useful tool for both object-centric as well as neural scene representation learning communities.

Author Information

Mehdi S. M. Sajjadi (Google)
Daniel Duckworth (Google Brain)
Aravindh Mahendran (Google)
Sjoerd van Steenkiste (Google Research)
Filip Pavetic (Google Switzerland GmbH)
Mario Lucic (Google Brain)
Leonidas Guibas (stanford.edu)
Klaus Greff (Google Brain)
Thomas Kipf (Google Research)

More from the Same Authors