Jun 10, 2019 · In this paper, we propose DensePhysNet, a system that actively executes a sequence of dynamic interactions (eg, sliding and colliding), and uses a deep ...
In this work, we propose to discover and learn the physical properties of objects through visual observations of multi- step, self-supervised, dynamic ...
The use of dense representation enables DensePhysNet to generalize well to novel scenes with more objects than in training. With knowledge of object physics, ...
DensePhysNet is proposed, a system that actively executes a sequence of dynamic interactions, and uses a deep predictive model over its visual observations ...
In this paper, we propose DensePhysNet, a system that actively executes a sequence of dynamic interactions (e.g., sliding and colliding), and uses a deep ...
Duration: 5:01
Posted: Jun 4, 2019
Posted: Jun 4, 2019
DensePhysNet: Learning Dense Physical Object Representations Via Multi-Step Dynamic Interactions. record by Zhenjia Xu • DensePhysNet: Learning Dense ...
Densephysnet: Learning dense physical object representations via multi-step dynamic interactions. Z Xu, J Wu, A Zeng, JB Tenenbaum, S Song. Robotics: Science ...
DensePhysNet: Learning Dense Physical Object Representations via Multi-Step Dynamic Interactions by Zhenjia Xu, Jiajun Wu, Andy Zeng, Joshua.
DensePhysNet: Learning Dense Physical Object Representations via Multi-step Dynamic Interactions · Unsupervised Discovery of Parts, Structure, and Dynamics.