Publication detail

Detection of facial key points

KANICH, O. GOLDMANN, T. DRAHANSKÝ, M.

Original Title

Detection of facial key points

Type

report

Language

English

Original Abstract

This report describes algorithm for detection of 98 facial key points using convolution neural nets (CNN). The neural nets use novel design based on state of the art design for similar tasks. WFLW (Wider Facial Landmarks in-the-wild) database which contains 10000 face images in atypical poses (illumination, makeup, expression, occlusion, etc.) is used for training and testing.

Authors

KANICH, O.; GOLDMANN, T.; DRAHANSKÝ, M.

Released

30. 12. 2019

Location

Brno

Pages from

1

Pages to

12

Pages count

12

URL

https://www.fit.vut.cz/research/publication/12160/

BibTex

@techreport{BUT162095,
  author="Ondřej {Kanich} and Tomáš {Goldmann} and Martin {Drahanský}",
  title="Detection of facial key points",
  year="2019",
  address="Brno",
  pages="1--12",
  url="https://www.fit.vut.cz/research/publication/12160/"
}

Documents

TZ_Keypoints.pdf