Submitted by Zhidong Liang.

Submission data

Full name3D Instance Embedding Learning With a Structure-Aware Loss Function
DescriptionStructure-Aware Loss Function for Point Cloud Segmentation

@ARTICLE{9126193,
author={Z. {Liang} and M. {Yang} and H. {Li} and C. {Wang}},
journal={IEEE Robotics and Automation Letters},
title={3D Instance Embedding Learning With a Structure-Aware Loss Function for Point Cloud Segmentation},
year={2020},
volume={5},
number={3},
pages={4915-4922},}
Publication title3D Instance Embedding Learning With a Structure-Aware Loss Function for Point Cloud Segmentation
Publication authorsZhidong Liang, Ming Yang, Hao Li, Chunxiang Wang
Publication venueIEEE Robotics and Automation Letters (IROS2020)
Publication URLhttps://ieeexplore.ieee.org/document/9126193?denied=
Input Data TypesUses Color,Uses Geometry        Uses 3D
Programming language(s)python
HardwareNVIDIA 1080TI
Submission creation date8 Dec, 2018
Last edited16 Feb, 2016

3D semantic instance results



Infoavg ap 25%bathtubbedbookshelfcabinetchaircountercurtaindeskdoorotherfurniturepicturerefrigeratorshower curtainsinksofatabletoiletwindow
0.6951.0000.8550.5790.5890.7350.4840.5880.8560.6340.5710.2980.5001.0000.8240.8180.7020.9350.545