Hair counting with deep learning

Alessio Gallucci; Dmitry Znamenskiy; Nicola Pezzotti; Milan Petkovic

doi:10.1109/BIA50171.2020.9244501

Hair counting with deep learning

Alessio Gallucci, Dmitry Znamenskiy, Nicola Pezzotti, Milan Petkovic

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

4 Citations (Scopus)

Abstract

We present a set of deep learning models aimed at solving the hair counting problem in human skin images. All the models are end-to-end, providing a mapping from the input image to a single scalar corresponding to the number of hair. The list of models corresponds to the most common deep learning architectures that worked over-time in various applications, where some of the networks were adapted to output the hair count. Results show that autoencoder architectures with skip connections work best for such end-to-end counting task, hinting at increased performance when multi-task learning is used. With the results presented, we speculate on the possibility to remove human annotator from the tedious task of manual counting of skin hair.

Original language	English
Title of host publication	Proceedings of the International Conference on Biomedical Innovations and Applications, BIA 2020
Editors	Valentina Markova, Todor Ganchev
Publisher	Institute of Electrical and Electronics Engineers
Pages	5-9
Number of pages	5
ISBN (Electronic)	9781728170732
DOIs	https://doi.org/10.1109/BIA50171.2020.9244501
Publication status	Published - 24 Sept 2020
Event	2020 International Conference on Biomedical Innovations and Applications, BIA 2020 - Varna, Bulgaria Duration: 24 Sept 2020 → 27 Sept 2020

Conference

Conference	2020 International Conference on Biomedical Innovations and Applications, BIA 2020
Country/Territory	Bulgaria
City	Varna
Period	24/09/20 → 27/09/20

Bibliographical note

Publisher Copyright:
© 2020 IEEE.

Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.

Keywords

computer vision
deep learning
dermatology
hair
object counting
skin

Access to Document

10.1109/BIA50171.2020.9244501

Cite this

@inproceedings{d45746763f8c4019837d8f8453d41109,

title = "Hair counting with deep learning",

abstract = "We present a set of deep learning models aimed at solving the hair counting problem in human skin images. All the models are end-to-end, providing a mapping from the input image to a single scalar corresponding to the number of hair. The list of models corresponds to the most common deep learning architectures that worked over-time in various applications, where some of the networks were adapted to output the hair count. Results show that autoencoder architectures with skip connections work best for such end-to-end counting task, hinting at increased performance when multi-task learning is used. With the results presented, we speculate on the possibility to remove human annotator from the tedious task of manual counting of skin hair. ",

keywords = "computer vision, deep learning, dermatology, hair, object counting, skin",

author = "Alessio Gallucci and Dmitry Znamenskiy and Nicola Pezzotti and Milan Petkovic",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE. Copyright: Copyright 2020 Elsevier B.V., All rights reserved.; 2020 International Conference on Biomedical Innovations and Applications, BIA 2020 ; Conference date: 24-09-2020 Through 27-09-2020",

year = "2020",

month = sep,

day = "24",

doi = "10.1109/BIA50171.2020.9244501",

language = "English",

pages = "5--9",

editor = "Valentina Markova and Todor Ganchev",

booktitle = "Proceedings of the International Conference on Biomedical Innovations and Applications, BIA 2020",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

}

Gallucci, A, Znamenskiy, D, Pezzotti, N & Petkovic, M 2020, Hair counting with deep learning. in V Markova & T Ganchev (eds), Proceedings of the International Conference on Biomedical Innovations and Applications, BIA 2020., 9244501, Institute of Electrical and Electronics Engineers, pp. 5-9, 2020 International Conference on Biomedical Innovations and Applications, BIA 2020, Varna, Bulgaria, 24/09/20. https://doi.org/10.1109/BIA50171.2020.9244501

Hair counting with deep learning. / Gallucci, Alessio; Znamenskiy, Dmitry; Pezzotti, Nicola et al.
Proceedings of the International Conference on Biomedical Innovations and Applications, BIA 2020. ed. / Valentina Markova; Todor Ganchev. Institute of Electrical and Electronics Engineers, 2020. p. 5-9 9244501.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Hair counting with deep learning

AU - Gallucci, Alessio

AU - Znamenskiy, Dmitry

AU - Pezzotti, Nicola

AU - Petkovic, Milan

PY - 2020/9/24

Y1 - 2020/9/24

N2 - We present a set of deep learning models aimed at solving the hair counting problem in human skin images. All the models are end-to-end, providing a mapping from the input image to a single scalar corresponding to the number of hair. The list of models corresponds to the most common deep learning architectures that worked over-time in various applications, where some of the networks were adapted to output the hair count. Results show that autoencoder architectures with skip connections work best for such end-to-end counting task, hinting at increased performance when multi-task learning is used. With the results presented, we speculate on the possibility to remove human annotator from the tedious task of manual counting of skin hair.

AB - We present a set of deep learning models aimed at solving the hair counting problem in human skin images. All the models are end-to-end, providing a mapping from the input image to a single scalar corresponding to the number of hair. The list of models corresponds to the most common deep learning architectures that worked over-time in various applications, where some of the networks were adapted to output the hair count. Results show that autoencoder architectures with skip connections work best for such end-to-end counting task, hinting at increased performance when multi-task learning is used. With the results presented, we speculate on the possibility to remove human annotator from the tedious task of manual counting of skin hair.

KW - computer vision

KW - deep learning

KW - dermatology

KW - hair

KW - object counting

KW - skin

UR - http://www.scopus.com/inward/record.url?scp=85096766683&partnerID=8YFLogxK

U2 - 10.1109/BIA50171.2020.9244501

DO - 10.1109/BIA50171.2020.9244501

M3 - Conference contribution

AN - SCOPUS:85096766683

SP - 5

EP - 9

BT - Proceedings of the International Conference on Biomedical Innovations and Applications, BIA 2020

A2 - Markova, Valentina

A2 - Ganchev, Todor

PB - Institute of Electrical and Electronics Engineers

T2 - 2020 International Conference on Biomedical Innovations and Applications, BIA 2020

Y2 - 24 September 2020 through 27 September 2020

ER -

Hair counting with deep learning

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this