Fast and Accurate Semantic Mapping through Geometric-based Incremental Segmentation

Yoshikatsu Nakajima, Keisuke Tateno, Federico Tombari, Hideo Saito

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

We propose an efficient and scalable method for incrementally building a dense, semantically annotated 3D map in real-time. The proposed method assigns class probabilities to each region, not each element (e.g., surfel and voxel), of the 3D map which is built up through a robust SLAM framework and incrementally segmented with a geometric-based segmentation method. Differently from all other approaches, our method has a capability of running at over 30Hz while performing all processing components, including SLAM, segmentation, 2D recognition, and updating class probabilities of each segmentation label at every incoming frame, thanks to the high efficiency that characterizes the computationally intensive stages of our framework. By utilizing a specifically designed CNN to improve the frame-wise segmentation result, we can also achieve high accuracy. We validate our method on the NYUv2 dataset by comparing with the state of the art in terms of accuracy and computational efficiency, and by means of an analysis in terms of time and space complexity.

Original languageEnglish
Title of host publication2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages385-392
Number of pages8
ISBN (Electronic)9781538680940
DOIs
Publication statusPublished - 2018 Dec 27
Event2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018 - Madrid, Spain
Duration: 2018 Oct 12018 Oct 5

Publication series

NameIEEE International Conference on Intelligent Robots and Systems
ISSN (Print)2153-0858
ISSN (Electronic)2153-0866

Conference

Conference2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018
CountrySpain
CityMadrid
Period18/10/118/10/5

Fingerprint

Semantics
Computational efficiency
Labels
Processing

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Software
  • Computer Vision and Pattern Recognition
  • Computer Science Applications

Cite this

Nakajima, Y., Tateno, K., Tombari, F., & Saito, H. (2018). Fast and Accurate Semantic Mapping through Geometric-based Incremental Segmentation. In 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018 (pp. 385-392). [8593993] (IEEE International Conference on Intelligent Robots and Systems). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IROS.2018.8593993

Fast and Accurate Semantic Mapping through Geometric-based Incremental Segmentation. / Nakajima, Yoshikatsu; Tateno, Keisuke; Tombari, Federico; Saito, Hideo.

2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 385-392 8593993 (IEEE International Conference on Intelligent Robots and Systems).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Nakajima, Y, Tateno, K, Tombari, F & Saito, H 2018, Fast and Accurate Semantic Mapping through Geometric-based Incremental Segmentation. in 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018., 8593993, IEEE International Conference on Intelligent Robots and Systems, Institute of Electrical and Electronics Engineers Inc., pp. 385-392, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018, Madrid, Spain, 18/10/1. https://doi.org/10.1109/IROS.2018.8593993
Nakajima Y, Tateno K, Tombari F, Saito H. Fast and Accurate Semantic Mapping through Geometric-based Incremental Segmentation. In 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 385-392. 8593993. (IEEE International Conference on Intelligent Robots and Systems). https://doi.org/10.1109/IROS.2018.8593993
Nakajima, Yoshikatsu ; Tateno, Keisuke ; Tombari, Federico ; Saito, Hideo. / Fast and Accurate Semantic Mapping through Geometric-based Incremental Segmentation. 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 385-392 (IEEE International Conference on Intelligent Robots and Systems).
@inproceedings{9b6ad1b6ac594edc851f5a2f742703ab,
title = "Fast and Accurate Semantic Mapping through Geometric-based Incremental Segmentation",
abstract = "We propose an efficient and scalable method for incrementally building a dense, semantically annotated 3D map in real-time. The proposed method assigns class probabilities to each region, not each element (e.g., surfel and voxel), of the 3D map which is built up through a robust SLAM framework and incrementally segmented with a geometric-based segmentation method. Differently from all other approaches, our method has a capability of running at over 30Hz while performing all processing components, including SLAM, segmentation, 2D recognition, and updating class probabilities of each segmentation label at every incoming frame, thanks to the high efficiency that characterizes the computationally intensive stages of our framework. By utilizing a specifically designed CNN to improve the frame-wise segmentation result, we can also achieve high accuracy. We validate our method on the NYUv2 dataset by comparing with the state of the art in terms of accuracy and computational efficiency, and by means of an analysis in terms of time and space complexity.",
author = "Yoshikatsu Nakajima and Keisuke Tateno and Federico Tombari and Hideo Saito",
year = "2018",
month = "12",
day = "27",
doi = "10.1109/IROS.2018.8593993",
language = "English",
series = "IEEE International Conference on Intelligent Robots and Systems",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "385--392",
booktitle = "2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018",

}

TY - GEN

T1 - Fast and Accurate Semantic Mapping through Geometric-based Incremental Segmentation

AU - Nakajima, Yoshikatsu

AU - Tateno, Keisuke

AU - Tombari, Federico

AU - Saito, Hideo

PY - 2018/12/27

Y1 - 2018/12/27

N2 - We propose an efficient and scalable method for incrementally building a dense, semantically annotated 3D map in real-time. The proposed method assigns class probabilities to each region, not each element (e.g., surfel and voxel), of the 3D map which is built up through a robust SLAM framework and incrementally segmented with a geometric-based segmentation method. Differently from all other approaches, our method has a capability of running at over 30Hz while performing all processing components, including SLAM, segmentation, 2D recognition, and updating class probabilities of each segmentation label at every incoming frame, thanks to the high efficiency that characterizes the computationally intensive stages of our framework. By utilizing a specifically designed CNN to improve the frame-wise segmentation result, we can also achieve high accuracy. We validate our method on the NYUv2 dataset by comparing with the state of the art in terms of accuracy and computational efficiency, and by means of an analysis in terms of time and space complexity.

AB - We propose an efficient and scalable method for incrementally building a dense, semantically annotated 3D map in real-time. The proposed method assigns class probabilities to each region, not each element (e.g., surfel and voxel), of the 3D map which is built up through a robust SLAM framework and incrementally segmented with a geometric-based segmentation method. Differently from all other approaches, our method has a capability of running at over 30Hz while performing all processing components, including SLAM, segmentation, 2D recognition, and updating class probabilities of each segmentation label at every incoming frame, thanks to the high efficiency that characterizes the computationally intensive stages of our framework. By utilizing a specifically designed CNN to improve the frame-wise segmentation result, we can also achieve high accuracy. We validate our method on the NYUv2 dataset by comparing with the state of the art in terms of accuracy and computational efficiency, and by means of an analysis in terms of time and space complexity.

UR - http://www.scopus.com/inward/record.url?scp=85062952472&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062952472&partnerID=8YFLogxK

U2 - 10.1109/IROS.2018.8593993

DO - 10.1109/IROS.2018.8593993

M3 - Conference contribution

T3 - IEEE International Conference on Intelligent Robots and Systems

SP - 385

EP - 392

BT - 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -