On-the-fly detection of novel objects in indoor environments

Edith Langer, Bram Ridder, Michael Cashmore, Daniele Magazzeni, Michael Zillich, Markus Vincze

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

Abstract

Many robotic applications require the detection of new objects in known environments. Common approaches navigate in the environment using pre-defined waypoints and segment the scene at these waypoints. Without knowing where to find new objects, this process can be time-consuming and prone to detecting false positives. To overcome these limitations we propose an approach that combines navigation and attention in order to detect novel objects rapidly. We exploit the octomap, created by the robot while it navigates in the environment, as a pre-attention filter to suggest potential regions of interest. These regions are then visited to obtain a close-up view for better object detection and recognition. We evaluate our approach in a simulated as well as a real environment. The experiments show that our approach outperforms previous approaches in terms of runtime and the number of segmentation actions required to find all novel objects in the environment.
Original languageEnglish
Title of host publication2017 IEEE International Conference on Robotics and Biomimetics, ROBIO 2017
Place of PublicationPiscataway, NJ.
PublisherIEEE
Number of pages8
Volume2018-January
ISBN (Print)9781538637418
DOIs
Publication statusPublished - 26 Mar 2018
Event2017 IEEE International Conference on Robotics and Biomimetics (ROBIO) - Macau, China
Duration: 5 Dec 20178 Dec 2017
Conference number: ROBIO2017

Conference

Conference2017 IEEE International Conference on Robotics and Biomimetics (ROBIO)
CountryChina
CityMacau
Period5/12/178/12/17

Fingerprint

Object recognition
Navigation
Robotics
Robots
Experiments
Object detection

Keywords

  • robot sensing systems
  • navigation
  • three-dimensional displays
  • task analysis
  • search problems
  • object recognition
  • collision avoidance
  • image filtering
  • image segmentation
  • mobile robots
  • object detection
  • robot vision

Cite this

Langer, E., Ridder, B., Cashmore, M., Magazzeni, D., Zillich, M., & Vincze, M. (2018). On-the-fly detection of novel objects in indoor environments. In 2017 IEEE International Conference on Robotics and Biomimetics, ROBIO 2017 (Vol. 2018-January). Piscataway, NJ.: IEEE. https://doi.org/10.1109/ROBIO.2017.8324532
Langer, Edith ; Ridder, Bram ; Cashmore, Michael ; Magazzeni, Daniele ; Zillich, Michael ; Vincze, Markus. / On-the-fly detection of novel objects in indoor environments. 2017 IEEE International Conference on Robotics and Biomimetics, ROBIO 2017. Vol. 2018-January Piscataway, NJ. : IEEE, 2018.
@inproceedings{9b1d72f3e605459faa7822ac3abc6137,
title = "On-the-fly detection of novel objects in indoor environments",
abstract = "Many robotic applications require the detection of new objects in known environments. Common approaches navigate in the environment using pre-defined waypoints and segment the scene at these waypoints. Without knowing where to find new objects, this process can be time-consuming and prone to detecting false positives. To overcome these limitations we propose an approach that combines navigation and attention in order to detect novel objects rapidly. We exploit the octomap, created by the robot while it navigates in the environment, as a pre-attention filter to suggest potential regions of interest. These regions are then visited to obtain a close-up view for better object detection and recognition. We evaluate our approach in a simulated as well as a real environment. The experiments show that our approach outperforms previous approaches in terms of runtime and the number of segmentation actions required to find all novel objects in the environment.",
keywords = "robot sensing systems, navigation, three-dimensional displays, task analysis, search problems, object recognition, collision avoidance, image filtering, image segmentation, mobile robots, object detection, robot vision",
author = "Edith Langer and Bram Ridder and Michael Cashmore and Daniele Magazzeni and Michael Zillich and Markus Vincze",
year = "2018",
month = "3",
day = "26",
doi = "10.1109/ROBIO.2017.8324532",
language = "English",
isbn = "9781538637418",
volume = "2018-January",
booktitle = "2017 IEEE International Conference on Robotics and Biomimetics, ROBIO 2017",
publisher = "IEEE",

}

Langer, E, Ridder, B, Cashmore, M, Magazzeni, D, Zillich, M & Vincze, M 2018, On-the-fly detection of novel objects in indoor environments. in 2017 IEEE International Conference on Robotics and Biomimetics, ROBIO 2017. vol. 2018-January, IEEE, Piscataway, NJ., 2017 IEEE International Conference on Robotics and Biomimetics (ROBIO), Macau, China, 5/12/17. https://doi.org/10.1109/ROBIO.2017.8324532

On-the-fly detection of novel objects in indoor environments. / Langer, Edith; Ridder, Bram; Cashmore, Michael ; Magazzeni, Daniele ; Zillich, Michael; Vincze, Markus.

2017 IEEE International Conference on Robotics and Biomimetics, ROBIO 2017. Vol. 2018-January Piscataway, NJ. : IEEE, 2018.

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

TY - GEN

T1 - On-the-fly detection of novel objects in indoor environments

AU - Langer, Edith

AU - Ridder, Bram

AU - Cashmore, Michael

AU - Magazzeni, Daniele

AU - Zillich, Michael

AU - Vincze, Markus

PY - 2018/3/26

Y1 - 2018/3/26

N2 - Many robotic applications require the detection of new objects in known environments. Common approaches navigate in the environment using pre-defined waypoints and segment the scene at these waypoints. Without knowing where to find new objects, this process can be time-consuming and prone to detecting false positives. To overcome these limitations we propose an approach that combines navigation and attention in order to detect novel objects rapidly. We exploit the octomap, created by the robot while it navigates in the environment, as a pre-attention filter to suggest potential regions of interest. These regions are then visited to obtain a close-up view for better object detection and recognition. We evaluate our approach in a simulated as well as a real environment. The experiments show that our approach outperforms previous approaches in terms of runtime and the number of segmentation actions required to find all novel objects in the environment.

AB - Many robotic applications require the detection of new objects in known environments. Common approaches navigate in the environment using pre-defined waypoints and segment the scene at these waypoints. Without knowing where to find new objects, this process can be time-consuming and prone to detecting false positives. To overcome these limitations we propose an approach that combines navigation and attention in order to detect novel objects rapidly. We exploit the octomap, created by the robot while it navigates in the environment, as a pre-attention filter to suggest potential regions of interest. These regions are then visited to obtain a close-up view for better object detection and recognition. We evaluate our approach in a simulated as well as a real environment. The experiments show that our approach outperforms previous approaches in terms of runtime and the number of segmentation actions required to find all novel objects in the environment.

KW - robot sensing systems

KW - navigation

KW - three-dimensional displays

KW - task analysis

KW - search problems

KW - object recognition

KW - collision avoidance

KW - image filtering

KW - image segmentation

KW - mobile robots

KW - object detection

KW - robot vision

U2 - 10.1109/ROBIO.2017.8324532

DO - 10.1109/ROBIO.2017.8324532

M3 - Conference contribution book

SN - 9781538637418

VL - 2018-January

BT - 2017 IEEE International Conference on Robotics and Biomimetics, ROBIO 2017

PB - IEEE

CY - Piscataway, NJ.

ER -

Langer E, Ridder B, Cashmore M, Magazzeni D, Zillich M, Vincze M. On-the-fly detection of novel objects in indoor environments. In 2017 IEEE International Conference on Robotics and Biomimetics, ROBIO 2017. Vol. 2018-January. Piscataway, NJ.: IEEE. 2018 https://doi.org/10.1109/ROBIO.2017.8324532