A 3D descriptor to detect task-oriented grasping points in clothing

作者：

Highlights：

• Novel 3D descriptor for manipulation-oriented perception of garments.

• Two orders of magnitude faster than state-of-the-art descriptors with performance.

• Select a grasping point for a garment manipulation task (e.g., hanging a shirt).

• Coarse-to-fine approach that significantly increases accuracy and close to real time.

摘要

Highlights•Novel 3D descriptor for manipulation-oriented perception of garments.•Two orders of magnitude faster than state-of-the-art descriptors with performance.•Select a grasping point for a garment manipulation task (e.g., hanging a shirt).•Coarse-to-fine approach that significantly increases accuracy and close to real time.

论文关键词：3D descriptor,Recognition,Detection,Grasping,Manipulation,Robotics

论文评审过程：Received 13 April 2015, Revised 11 May 2016, Accepted 3 July 2016, Available online 6 July 2016, Version of Record 26 July 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.07.003