• Publications
  • Influence
Shape completion enabled robotic grasping
TLDR
We train a CNN to complete and mesh an object observed from a single point of view, and then plan grasps on the completed object. Expand
  • 124
  • 11
  • PDF
Generating multi-fingered robotic grasps via deep learning
TLDR
This paper presents a deep learning architecture for detecting the palm and fingertip positions of stable grasps directly from partial object views. Expand
  • 67
  • 4
  • PDF
Workspace Aware Online Grasp Planning
TLDR
This work provides a framework for a workspace aware online grasp planner. Expand
  • 5
  • 1
  • PDF
Learning To Grasp
  • 2
  • 1
Multi-Modal Geometric Learning for Grasping and Manipulation
TLDR
This work provides an architecture that incorporates depth and tactile information to create rich and accurate 3D models useful for robotic manipulation tasks. Expand
  • 22
  • PDF
MAT: Multi-Fingered Adaptive Tactile Grasping via Deep Reinforcement Learning
TLDR
We present Multi-Fingered Adaptive Tactile Grasping, or MAT, a tactile closed-loop method capable of realizing grasps provided by a coarse initial positioning of the hand above an object. Expand
  • 9
  • PDF
Generative Attention Learning: a "GenerAL" framework for high-performance multi-fingered grasping in clutter
TLDR
Generative Attention Learning (GenerAL) is a framework for high-DOF multi-fingered grasping that is not only robust to dense clutter and novel objects but also effective with a variety of different parallel-jaw and multi-fingerered robot hands. Expand
  • 3
  • PDF
Learning Precise 3D Manipulation from Multiple Uncalibrated Cameras
TLDR
We present an effective multi-view approach to closed-loop end-to-end learning of precise manipulation tasks that are 3D in nature. Expand
  • 3
  • PDF
An Ode to an ODE
TLDR
We present a new paradigm for Neural ODE algorithms, called ODEtoODE, where time-dependent parameters of the main flow evolve according to a matrix flow on the orthogonal group O(d). Expand
  • 1
  • PDF