• Publications
  • Influence
VAIS: A dataset for recognizing maritime imagery in the visible and infrared spectrums
This paper introduces VAIS, the world's first publicly available dataset of paired visible and infrared ship imagery. Expand
Single image 3D object detection and pose estimation for grasping
We present a novel approach for detecting objects and estimating their 3D pose in single images of cluttered scenes using a deformable parts-based model. Expand
A triangle histogram for object classification by tactile sensing
We present a new descriptor for tactile 3D object classification. Expand
Tactile Perception And Visuotactile Integration For Robotic Exploration
We approach the problem of perception prior to manipulation, using the sense of touch, for the purpose of understanding the surroundings of an autonomous robot. Expand
Tactile-Visual Integration for Task-Aware
Related Work Varley, Watkins-Valls, and Allen. Multi-Modal Geometric Learning for Grasping and Manipulation. arXiv 2018.! Calandra, Owens, Upadhyaya, Yuan, Lin, Adelson, Levine. The feeling ofExpand
Active end-effector pose selection for tactile object recognition through Monte Carlo tree search
This paper considers the problem of active object recognition using touch only. Expand
Shape-based object classification and recognition through continuum manipulation
We introduce a novel approach to shape-based object classification and recognition through the use of a continuum manipulator. Expand
Tactile-Vision Integration for Task-Compatible Fine-Part Manipulation
We propose to integrate tactile and visual sensing to predict task-compatible grasp regions for manipulation. We address the problem of fine-part assembly, by leveraging vision to observe scene-levelExpand
Disambiguation of Horizontal Direction for Video Conference Systems
We present a video tele-conferencing concept, which, by combining existing software and hardware technologies, achieves horizontal directionality for multiple sites and participants at each site. Expand
Necessity for More Realistic Contact Simulation
Given the recent progress in large-scale data-driven methods, sim-to-real is quickly becoming an enabler for access to machine learning research, alleviating hardware costs and physical constraints.Expand