Timezone: »

 
Poster
Unsupervised Learning of 3D Structure from Images
Danilo Jimenez Rezende · S. M. Ali Eslami · Shakir Mohamed · Peter Battaglia · Max Jaderberg · Nicolas Heess

Wed Dec 07 09:00 AM -- 12:30 PM (PST) @ Area 5+6+7+8 #2

A key goal of computer vision is to recover the underlying 3D structure that gives rise to 2D observations of the world. If endowed with 3D understanding, agents can abstract away from the complexity of the rendering process to form stable, disentangled representations of scene elements. In this paper we learn strong deep generative models of 3D structures, and recover these structures from 2D images via probabilistic inference. We demonstrate high-quality samples and report log-likelihoods on several datasets, including ShapeNet, and establish the first benchmarks in the literature. We also show how these models and their inference networks can be trained jointly, end-to-end, and directly from 2D images without any use of ground-truth 3D labels. This demonstrates for the first time the feasibility of learning to infer 3D representations of the world in a purely unsupervised manner.

Author Information

Danilo Jimenez Rezende (Google DeepMind)
S. M. Ali Eslami (Google DeepMind)
Shakir Mohamed (Google DeepMind)
Shakir Mohamed

Shakir Mohamed is a senior staff scientist at DeepMind in London. Shakir's main interests lie at the intersection of approximate Bayesian inference, deep learning and reinforcement learning, and the role that machine learning systems at this intersection have in the development of more intelligent and general-purpose learning systems. Before moving to London, Shakir held a Junior Research Fellowship from the Canadian Institute for Advanced Research (CIFAR), based in Vancouver at the University of British Columbia with Nando de Freitas. Shakir completed his PhD with Zoubin Ghahramani at the University of Cambridge, where he was a Commonwealth Scholar to the United Kingdom. Shakir is from South Africa and completed his previous degrees in Electrical and Information Engineering at the University of the Witwatersrand, Johannesburg.

Peter Battaglia (Google DeepMind)
Max Jaderberg (Isomorphic Labs)
Nicolas Heess (Google DeepMind)

More from the Same Authors