Discriminative feature learning for efficient RGB-D object recognition

Research output: Chapter in Book/Conference paperConference paper

8 Citations (Scopus)

Abstract

© 2015 IEEE. This paper presents an efficient approach to recognize objects captured with an RGB-D sensor. The proposed approach uses a Bag-of-Words (BOW) model to learn feature representations from raw RGB-D point clouds in a weakly supervised manner. To this end, we introduce a novel method based on randomized clustering trees to learn visual vocabularies which are fast to compute and more discriminative compared to the vocabularies generated by classical methods such as k-means. We show that, when combined with standard spatial pooling strategies, our proposed approach yields a powerful feature representation for RGB-D object recognition. Our extensive experimental evaluation on two challenging RGB-D object datasets and live video streams from Kinect shows that our learned features result in superior object recognition accuracies compared with the state-of-the-art methods.
Original languageEnglish
Title of host publicationIEEE International Conference on Intelligent Robots and Systems
Place of PublicationUnited States
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages272-279
Volume2015
ISBN (Print)9781479999941
DOIs
Publication statusPublished - 2015
Event2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) - Congress Center Hamburg Am Dammtor / Marseiller Straße, Hamburg, Germany
Duration: 28 Sep 20152 Oct 2015
http://www.iros2015.org

Conference

Conference2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
CountryGermany
CityHamburg
Period28/09/152/10/15
Internet address

Fingerprint

Object recognition
Sensors

Cite this

Asif, U., Bennamoun, M., & Sohel, F. (2015). Discriminative feature learning for efficient RGB-D object recognition. In IEEE International Conference on Intelligent Robots and Systems (Vol. 2015, pp. 272-279). United States: IEEE, Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/IROS.2015.7353385
Asif, Umar ; Bennamoun, Mohammed ; Sohel, Ferdous. / Discriminative feature learning for efficient RGB-D object recognition. IEEE International Conference on Intelligent Robots and Systems. Vol. 2015 United States : IEEE, Institute of Electrical and Electronics Engineers, 2015. pp. 272-279
@inproceedings{8dd887bec763448dbe8fb04b661c9ff2,
title = "Discriminative feature learning for efficient RGB-D object recognition",
abstract = "{\circledC} 2015 IEEE. This paper presents an efficient approach to recognize objects captured with an RGB-D sensor. The proposed approach uses a Bag-of-Words (BOW) model to learn feature representations from raw RGB-D point clouds in a weakly supervised manner. To this end, we introduce a novel method based on randomized clustering trees to learn visual vocabularies which are fast to compute and more discriminative compared to the vocabularies generated by classical methods such as k-means. We show that, when combined with standard spatial pooling strategies, our proposed approach yields a powerful feature representation for RGB-D object recognition. Our extensive experimental evaluation on two challenging RGB-D object datasets and live video streams from Kinect shows that our learned features result in superior object recognition accuracies compared with the state-of-the-art methods.",
author = "Umar Asif and Mohammed Bennamoun and Ferdous Sohel",
year = "2015",
doi = "10.1109/IROS.2015.7353385",
language = "English",
isbn = "9781479999941",
volume = "2015",
pages = "272--279",
booktitle = "IEEE International Conference on Intelligent Robots and Systems",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",
address = "United States",

}

Asif, U, Bennamoun, M & Sohel, F 2015, Discriminative feature learning for efficient RGB-D object recognition. in IEEE International Conference on Intelligent Robots and Systems. vol. 2015, IEEE, Institute of Electrical and Electronics Engineers, United States, pp. 272-279, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Hamburg, Germany, 28/09/15. https://doi.org/10.1109/IROS.2015.7353385

Discriminative feature learning for efficient RGB-D object recognition. / Asif, Umar; Bennamoun, Mohammed; Sohel, Ferdous.

IEEE International Conference on Intelligent Robots and Systems. Vol. 2015 United States : IEEE, Institute of Electrical and Electronics Engineers, 2015. p. 272-279.

Research output: Chapter in Book/Conference paperConference paper

TY - GEN

T1 - Discriminative feature learning for efficient RGB-D object recognition

AU - Asif, Umar

AU - Bennamoun, Mohammed

AU - Sohel, Ferdous

PY - 2015

Y1 - 2015

N2 - © 2015 IEEE. This paper presents an efficient approach to recognize objects captured with an RGB-D sensor. The proposed approach uses a Bag-of-Words (BOW) model to learn feature representations from raw RGB-D point clouds in a weakly supervised manner. To this end, we introduce a novel method based on randomized clustering trees to learn visual vocabularies which are fast to compute and more discriminative compared to the vocabularies generated by classical methods such as k-means. We show that, when combined with standard spatial pooling strategies, our proposed approach yields a powerful feature representation for RGB-D object recognition. Our extensive experimental evaluation on two challenging RGB-D object datasets and live video streams from Kinect shows that our learned features result in superior object recognition accuracies compared with the state-of-the-art methods.

AB - © 2015 IEEE. This paper presents an efficient approach to recognize objects captured with an RGB-D sensor. The proposed approach uses a Bag-of-Words (BOW) model to learn feature representations from raw RGB-D point clouds in a weakly supervised manner. To this end, we introduce a novel method based on randomized clustering trees to learn visual vocabularies which are fast to compute and more discriminative compared to the vocabularies generated by classical methods such as k-means. We show that, when combined with standard spatial pooling strategies, our proposed approach yields a powerful feature representation for RGB-D object recognition. Our extensive experimental evaluation on two challenging RGB-D object datasets and live video streams from Kinect shows that our learned features result in superior object recognition accuracies compared with the state-of-the-art methods.

U2 - 10.1109/IROS.2015.7353385

DO - 10.1109/IROS.2015.7353385

M3 - Conference paper

SN - 9781479999941

VL - 2015

SP - 272

EP - 279

BT - IEEE International Conference on Intelligent Robots and Systems

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - United States

ER -

Asif U, Bennamoun M, Sohel F. Discriminative feature learning for efficient RGB-D object recognition. In IEEE International Conference on Intelligent Robots and Systems. Vol. 2015. United States: IEEE, Institute of Electrical and Electronics Engineers. 2015. p. 272-279 https://doi.org/10.1109/IROS.2015.7353385