Hair counting with deep learning

Alessio Gallucci, Dmitry Znamenskiy, Nicola Pezzotti, Milan Petkovic

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

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 languageEnglish
Title of host publicationProceedings of the International Conference on Biomedical Innovations and Applications, BIA 2020
EditorsValentina Markova, Todor Ganchev
PublisherInstitute of Electrical and Electronics Engineers
Pages5-9
Number of pages5
ISBN (Electronic)9781728170732
DOIs
Publication statusPublished - 24 Sep 2020
Event2020 International Conference on Biomedical Innovations and Applications, BIA 2020 - Varna, Bulgaria
Duration: 24 Sep 202027 Sep 2020

Conference

Conference2020 International Conference on Biomedical Innovations and Applications, BIA 2020
CountryBulgaria
CityVarna
Period24/09/2027/09/20

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

Fingerprint Dive into the research topics of 'Hair counting with deep learning'. Together they form a unique fingerprint.

Cite this