Multi-label Fashion Image Classification with Minimal Human Supervision

Naoto Inoue, Edgar Simo Serra, Toshihiko Yamasaki, Hiroshi Ishikawa

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

7 Citations (Scopus)

Abstract

We tackle the problem of multi-label classification of fashion images, learning from noisy data with minimal human supervision. We present a new dataset of full body poses, each with a set of 66 binary labels corresponding to the information about the garments worn in the image obtained in an automatic manner. As the automatically-collected labels contain significant noise, we manually correct the labels for a small subset of the data, and use these correct labels for further training and evaluation. We build upon a recent approach that both cleans the noisy labels and learns to classify, and introduce simple changes that can significantly improve the performance.

Original languageEnglish
Title of host publicationProceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages2261-2267
Number of pages7
Volume2018-January
ISBN (Electronic)9781538610343
DOIs
Publication statusPublished - 2018 Jan 19
Event16th IEEE International Conference on Computer Vision Workshops, ICCVW 2017 - Venice, Italy
Duration: 2017 Oct 222017 Oct 29

Other

Other16th IEEE International Conference on Computer Vision Workshops, ICCVW 2017
CountryItaly
CityVenice
Period17/10/2217/10/29

Fingerprint

Image classification
Labels

ASJC Scopus subject areas

  • Computer Science Applications
  • Computer Vision and Pattern Recognition

Cite this

Inoue, N., Simo Serra, E., Yamasaki, T., & Ishikawa, H. (2018). Multi-label Fashion Image Classification with Minimal Human Supervision. In Proceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017 (Vol. 2018-January, pp. 2261-2267). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCVW.2017.265

Multi-label Fashion Image Classification with Minimal Human Supervision. / Inoue, Naoto; Simo Serra, Edgar; Yamasaki, Toshihiko; Ishikawa, Hiroshi.

Proceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017. Vol. 2018-January Institute of Electrical and Electronics Engineers Inc., 2018. p. 2261-2267.

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

Inoue, N, Simo Serra, E, Yamasaki, T & Ishikawa, H 2018, Multi-label Fashion Image Classification with Minimal Human Supervision. in Proceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017. vol. 2018-January, Institute of Electrical and Electronics Engineers Inc., pp. 2261-2267, 16th IEEE International Conference on Computer Vision Workshops, ICCVW 2017, Venice, Italy, 17/10/22. https://doi.org/10.1109/ICCVW.2017.265
Inoue N, Simo Serra E, Yamasaki T, Ishikawa H. Multi-label Fashion Image Classification with Minimal Human Supervision. In Proceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017. Vol. 2018-January. Institute of Electrical and Electronics Engineers Inc. 2018. p. 2261-2267 https://doi.org/10.1109/ICCVW.2017.265
Inoue, Naoto ; Simo Serra, Edgar ; Yamasaki, Toshihiko ; Ishikawa, Hiroshi. / Multi-label Fashion Image Classification with Minimal Human Supervision. Proceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017. Vol. 2018-January Institute of Electrical and Electronics Engineers Inc., 2018. pp. 2261-2267
@inproceedings{62d5de18db394009a79482a3a3a3abe9,
title = "Multi-label Fashion Image Classification with Minimal Human Supervision",
abstract = "We tackle the problem of multi-label classification of fashion images, learning from noisy data with minimal human supervision. We present a new dataset of full body poses, each with a set of 66 binary labels corresponding to the information about the garments worn in the image obtained in an automatic manner. As the automatically-collected labels contain significant noise, we manually correct the labels for a small subset of the data, and use these correct labels for further training and evaluation. We build upon a recent approach that both cleans the noisy labels and learns to classify, and introduce simple changes that can significantly improve the performance.",
author = "Naoto Inoue and {Simo Serra}, Edgar and Toshihiko Yamasaki and Hiroshi Ishikawa",
year = "2018",
month = "1",
day = "19",
doi = "10.1109/ICCVW.2017.265",
language = "English",
volume = "2018-January",
pages = "2261--2267",
booktitle = "Proceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Multi-label Fashion Image Classification with Minimal Human Supervision

AU - Inoue, Naoto

AU - Simo Serra, Edgar

AU - Yamasaki, Toshihiko

AU - Ishikawa, Hiroshi

PY - 2018/1/19

Y1 - 2018/1/19

N2 - We tackle the problem of multi-label classification of fashion images, learning from noisy data with minimal human supervision. We present a new dataset of full body poses, each with a set of 66 binary labels corresponding to the information about the garments worn in the image obtained in an automatic manner. As the automatically-collected labels contain significant noise, we manually correct the labels for a small subset of the data, and use these correct labels for further training and evaluation. We build upon a recent approach that both cleans the noisy labels and learns to classify, and introduce simple changes that can significantly improve the performance.

AB - We tackle the problem of multi-label classification of fashion images, learning from noisy data with minimal human supervision. We present a new dataset of full body poses, each with a set of 66 binary labels corresponding to the information about the garments worn in the image obtained in an automatic manner. As the automatically-collected labels contain significant noise, we manually correct the labels for a small subset of the data, and use these correct labels for further training and evaluation. We build upon a recent approach that both cleans the noisy labels and learns to classify, and introduce simple changes that can significantly improve the performance.

UR - http://www.scopus.com/inward/record.url?scp=85046289547&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046289547&partnerID=8YFLogxK

U2 - 10.1109/ICCVW.2017.265

DO - 10.1109/ICCVW.2017.265

M3 - Conference contribution

AN - SCOPUS:85046289547

VL - 2018-January

SP - 2261

EP - 2267

BT - Proceedings - 2017 IEEE International Conference on Computer Vision Workshops, ICCVW 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -