• Publications
  • Influence
An In Depth View of Saliency
TLDR
We propose a novel method which incorporates depth measurements into the computation of visual saliency. Expand
  • 74
  • 11
  • PDF
Movie genre classification via scene categorization
TLDR
This paper presents a method for movie genre categorization of movie trailers, based on scene categorization. Expand
  • 91
  • 5
  • PDF
Affordance Prediction via Learned Object Attributes
We present a novel method for learning and predicting the affordances of an object based on its physical and visual attributes. Affordance prediction is a key task in autonomous robot learning, as itExpand
  • 64
  • 5
  • PDF
Planning Multi-Fingered Grasps as Probabilistic Inference in a Learned Deep Network
TLDR
We propose a novel approach to multi-fingered grasp planning leveraging learned deep neural network models. Expand
  • 39
  • 4
  • PDF
Decoupling behavior, perception, and control for autonomous learning of affordances
TLDR
A novel behavior representation is introduced that permits a robot to systematically explore the best methods by which to successfully execute an affordance-based behavior for a particular object. Expand
  • 28
  • 4
  • PDF
Learning contact locations for pushing and orienting unknown objects
TLDR
We present a method by which a robot learns to predict effective contact locations for pushing as a function of object shape, regardless of whether they belong to a previously encountered object class. Expand
  • 39
  • 3
  • PDF
Learning robot in-hand manipulation with tactile features
TLDR
We propose a reinforcement learning method to learn a tactile skill on a compliant, under-actuated robot hand that generalizes to novel objects. Expand
  • 97
  • 2
  • PDF
Stabilizing novel objects by learning to predict tactile slip
TLDR
We explore the generalization capabilities of well known supervised learning methods, using random forest classifiers to create generalizable slip predictors. Expand
  • 58
  • 2
  • PDF
Active tactile object exploration with Gaussian processes
TLDR
We present an active touch strategy to efficiently reduce the surface geometry uncertainty for tactile object surface modeling by leveraging a probabilistic representation of object surface. Expand
  • 42
  • 2
  • PDF
Robust Learning of Tactile Force Estimation through Robot Interaction
TLDR
We explore learning a robust model that maps tactile sensor signals to force via neural networks. Expand
  • 19
  • 2
  • PDF