Generative OpenMax for multi-class open set classification

Zongyuan Ge, Sergey Demyanov, Zetao Chen, Rahil Garnavi

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

1 Citation (Scopus)

Abstract

We present a conceptually new and flexible method for multi-class open set classification. Unlike previous methods where unknown classes are inferred with respect to the feature or decision distance to the known classes, our approach is able to provide explicit modelling and decision score for unknown classes. The proposed method, called Generative OpenMax (G-OpenMax), extends OpenMax by employing generative adversarial networks (GANs) for novel category image synthesis. We validate the proposed method on two datasets of handwritten digits and characters, resulting in superior results over previous deep learning based method OpenMax Moreover, G-OpenMax provides a way to visualize samples representing the unknown classes from open space.
Original languageEnglish
Title of host publicationBritish Machine Vision Conference Proceedings 2017
EditorsKrystian Mikolajczyk, Gabriel Brostow
Place of PublicationLondon UK
PublisherBritish Machine Vision Association and Society for Pattern Recognition
Number of pages12
ISBN (Electronic)9781901725605
ISBN (Print)190172560X
DOIs
Publication statusPublished - 2017
Externally publishedYes
EventBritish Machine Vision Conference 2017 - Royal Geographic Society of London, London, United Kingdom
Duration: 4 Sep 20177 Sep 2017
Conference number: 28th
https://bmvc2017.london/

Conference

ConferenceBritish Machine Vision Conference 2017
Abbreviated titleBMVC 2017
CountryUnited Kingdom
CityLondon
Period4/09/177/09/17
Internet address

Cite this

Ge, Z., Demyanov, S., Chen, Z., & Garnavi, R. (2017). Generative OpenMax for multi-class open set classification. In K. Mikolajczyk, & G. Brostow (Eds.), British Machine Vision Conference Proceedings 2017 London UK: British Machine Vision Association and Society for Pattern Recognition. https://doi.org/10.5244/C.31.42
Ge, Zongyuan ; Demyanov, Sergey ; Chen, Zetao ; Garnavi, Rahil. / Generative OpenMax for multi-class open set classification. British Machine Vision Conference Proceedings 2017. editor / Krystian Mikolajczyk ; Gabriel Brostow. London UK : British Machine Vision Association and Society for Pattern Recognition, 2017.
@inproceedings{5d2137fb17cf43b683e7d4225acc6a9b,
title = "Generative OpenMax for multi-class open set classification",
abstract = "We present a conceptually new and flexible method for multi-class open set classification. Unlike previous methods where unknown classes are inferred with respect to the feature or decision distance to the known classes, our approach is able to provide explicit modelling and decision score for unknown classes. The proposed method, called Generative OpenMax (G-OpenMax), extends OpenMax by employing generative adversarial networks (GANs) for novel category image synthesis. We validate the proposed method on two datasets of handwritten digits and characters, resulting in superior results over previous deep learning based method OpenMax Moreover, G-OpenMax provides a way to visualize samples representing the unknown classes from open space.",
author = "Zongyuan Ge and Sergey Demyanov and Zetao Chen and Rahil Garnavi",
year = "2017",
doi = "10.5244/C.31.42",
language = "English",
isbn = "190172560X",
editor = "Krystian Mikolajczyk and Gabriel Brostow",
booktitle = "British Machine Vision Conference Proceedings 2017",
publisher = "British Machine Vision Association and Society for Pattern Recognition",

}

Ge, Z, Demyanov, S, Chen, Z & Garnavi, R 2017, Generative OpenMax for multi-class open set classification. in K Mikolajczyk & G Brostow (eds), British Machine Vision Conference Proceedings 2017. British Machine Vision Association and Society for Pattern Recognition, London UK, British Machine Vision Conference 2017, London, United Kingdom, 4/09/17. https://doi.org/10.5244/C.31.42

Generative OpenMax for multi-class open set classification. / Ge, Zongyuan; Demyanov, Sergey; Chen, Zetao; Garnavi, Rahil.

British Machine Vision Conference Proceedings 2017. ed. / Krystian Mikolajczyk; Gabriel Brostow. London UK : British Machine Vision Association and Society for Pattern Recognition, 2017.

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

TY - GEN

T1 - Generative OpenMax for multi-class open set classification

AU - Ge, Zongyuan

AU - Demyanov, Sergey

AU - Chen, Zetao

AU - Garnavi, Rahil

PY - 2017

Y1 - 2017

N2 - We present a conceptually new and flexible method for multi-class open set classification. Unlike previous methods where unknown classes are inferred with respect to the feature or decision distance to the known classes, our approach is able to provide explicit modelling and decision score for unknown classes. The proposed method, called Generative OpenMax (G-OpenMax), extends OpenMax by employing generative adversarial networks (GANs) for novel category image synthesis. We validate the proposed method on two datasets of handwritten digits and characters, resulting in superior results over previous deep learning based method OpenMax Moreover, G-OpenMax provides a way to visualize samples representing the unknown classes from open space.

AB - We present a conceptually new and flexible method for multi-class open set classification. Unlike previous methods where unknown classes are inferred with respect to the feature or decision distance to the known classes, our approach is able to provide explicit modelling and decision score for unknown classes. The proposed method, called Generative OpenMax (G-OpenMax), extends OpenMax by employing generative adversarial networks (GANs) for novel category image synthesis. We validate the proposed method on two datasets of handwritten digits and characters, resulting in superior results over previous deep learning based method OpenMax Moreover, G-OpenMax provides a way to visualize samples representing the unknown classes from open space.

U2 - 10.5244/C.31.42

DO - 10.5244/C.31.42

M3 - Conference Paper

SN - 190172560X

BT - British Machine Vision Conference Proceedings 2017

A2 - Mikolajczyk, Krystian

A2 - Brostow, Gabriel

PB - British Machine Vision Association and Society for Pattern Recognition

CY - London UK

ER -

Ge Z, Demyanov S, Chen Z, Garnavi R. Generative OpenMax for multi-class open set classification. In Mikolajczyk K, Brostow G, editors, British Machine Vision Conference Proceedings 2017. London UK: British Machine Vision Association and Society for Pattern Recognition. 2017 https://doi.org/10.5244/C.31.42