SegICP: Integrated Deep Semantic Segmentation and Pose Estimation

Recent robotic manipulation competitions have highlighted that sophisticated robots still struggle to achieve fast and reliable perception of task-relevant objects in complex, realistic scenarios. To improve these systems' perceptive speed and robustness, we present SegICP, a novel integrated s...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2017-09
Main Authors: Wong, Jay M, Kee, Vincent, Le, Tiffany, Wagner, Syler, Gian-Luca Mariottini, Schneider, Abraham, Hamilton, Lei, Chipalkatty, Rahul, Hebert, Mitchell, Johnson, David M S, Wu, Jimmy, Zhou, Bolei, Torralba, Antonio
Format: Article
Language:English
Subjects:
Artificial neural networks
Motion capture
Object recognition
Pose estimation
Position errors
Real time
Semantic segmentation
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Recent robotic manipulation competitions have highlighted that sophisticated robots still struggle to achieve fast and reliable perception of task-relevant objects in complex, realistic scenarios. To improve these systems' perceptive speed and robustness, we present SegICP, a novel integrated solution to object recognition and pose estimation. SegICP couples convolutional neural networks and multi-hypothesis point cloud registration to achieve both robust pixel-wise semantic segmentation as well as accurate and real-time 6-DOF pose estimation for relevant objects. Our architecture achieves 1cm position error and
ISSN:2331-8422
DOI:10.48550/arxiv.1703.01661