Timezone: »

Visual Grounding of Learned Physical Models
Yunzhu Li · Toru Lin · Kexin Yi · Daniel Bear · Daniel Yamins · Jiajun Wu · Josh Tenenbaum · Antonio Torralba

Thu Jul 16 08:00 AM -- 08:45 AM & Thu Jul 16 07:00 PM -- 07:45 PM (PDT) @ Virtual #None

Humans intuitively recognize objects' physical properties and predict their motion, even when the objects are engaged in complicated interactions. The abilities to perform physical reasoning and to adapt to new environments, while intrinsic to humans, remain challenging to state-of-the-art computational models. In this work, we present a neural model that simultaneously reasons about physics and makes future predictions based on visual and dynamics priors. The visual prior predicts a particle-based representation of the system from visual observations. An inference module operates on those particles, predicting and refining estimates of particle locations, object states, and physical parameters, subject to the constraints imposed by the dynamics prior, which we refer to as visual grounding. We demonstrate the effectiveness of our method in environments involving rigid objects, deformable materials, and fluids. Experiments show that our model can infer the physical properties within a few observations, which allows the model to quickly adapt to unseen scenarios and make accurate predictions into the future.

Author Information

Yunzhu Li (MIT)
Toru Lin (MIT)
Kexin Yi (Harvard University)
Daniel Bear (Stanford University)
Daniel Yamins (Stanford University)
Jiajun Wu (Stanford University)

Jiajun Wu is a Visiting Faculty Researcher at Google Research, New York City. In July 2020, he will join Stanford University as an Assistant Professor of Computer Science. He studies machine perception, reasoning, and its interaction with the physical world, drawing inspiration from human cognition.

Josh Tenenbaum (MIT)

Joshua Brett Tenenbaum is Professor of Cognitive Science and Computation at the Massachusetts Institute of Technology. He is known for contributions to mathematical psychology and Bayesian cognitive science. He previously taught at Stanford University, where he was the Wasow Visiting Fellow from October 2010 to January 2011. Tenenbaum received his undergraduate degree in physics from Yale University in 1993, and his Ph.D. from MIT in 1999. His work primarily focuses on analyzing probabilistic inference as the engine of human cognition and as a means to develop machine learning.

Antonio Torralba (MIT)

More from the Same Authors