An Approach to 3D Object Detection in Real-Time for Cognitive Robotics Experiments

Daniel Vidal-Soroa, Pedro Furelos, Francisco Bellas, José Antonio Becerra

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

This paper presents a computer vision method that, taking information from an RGB-D camera, performs real time 3D object recognition to be used in cognitive robotics experiments, where the real time constraints are key. To this end, we have implemented and tested an algorithm that combines a deep neural network (YOLOv3 tiny) that processes RGB images and provides object recognition and 2D localization, with a point cloud analysis method to compute the third dimension. The proposed approach has been tested in real-time manipulation experiments with the UR5e robotic arm through ROS, and using a GPU to execute the method, showing that this combination allows for an efficient real-time learning using cognitive models.

Original languageEnglish
Title of host publicationROBOT2022
Subtitle of host publication5th Iberian Robotics Conference - Advances in Robotics
EditorsDanilo Tardioli, Vicente Matellán, Guillermo Heredia, Manuel F. Silva, Lino Marques
PublisherSpringer Science and Business Media Deutschland GmbH
Pages283-294
Number of pages12
ISBN (Print)9783031210648
DOIs
StatePublished - 2023
Externally publishedYes
Event5th Iberian Robotics Conference, ROBOT 2022 - Zaragoza, Spain
Duration: 23 Nov 202225 Nov 2022

Publication series

NameLecture Notes in Networks and Systems
Volume589 LNNS
ISSN (Print)2367-3370
ISSN (Electronic)2367-3389

Conference

Conference5th Iberian Robotics Conference, ROBOT 2022
Country/TerritorySpain
CityZaragoza
Period23/11/2225/11/22

Keywords

  • 3D object detection
  • Cognitive Robotics
  • Computer vision
  • Deep learning
  • Real-time processing
  • RGB-D camera

Fingerprint

Dive into the research topics of 'An Approach to 3D Object Detection in Real-Time for Cognitive Robotics Experiments'. Together they form a unique fingerprint.

Cite this