SegICP: Integrated deep semantic segmentation and pose estimation

2017/03/31 by Jay Ming Wong, Jay M. Wong, Vincent Kee +13 · 151 citations
Computer Science · Engineering · #Advanced Neural Network Applications #Artificial intelligence #Benchmark (surveying) #Computer science #Computer vision #Convolutional neural network #Pixel #Point cloud #Pose #Robot #Robot Manipulation and Learning #Robotics and Sensor-Based Localization #Robustness (evolution) #Segmentation #cs.CV #cs.RO

paper · pdf · doi:10.1109/iros.2017.8206470

IROS camera-ready

openalex publication_date 2017/09/01 · arxiv created 2017/09/05 · arxiv updated 2018/03/13 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05

Abstract

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 1 cm position error and <; 5° angle error in real time without an initial seed. We evaluate and benchmark SegICP against an annotated dataset generated by motion capture.

Citations

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• LabelMe: A Database and Web-Based Tool for Image Annotation
• Gradient-based learning applied to document recognition
• Distinctive Image Features from Scale-Invariant Keypoints
• ImageNet classification with deep convolutional neural networks
• Semantic Understanding of Scenes Through the ADE20K Dataset
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal\n Networks
• Semantic Understanding of Scenes through the ADE20K Dataset

Cited by

• TriVoC: Efficient Voting-based Consensus Maximization for Robust Point Cloud Registration with Extreme Outlier Ratios