A Multi-Task Bayesian Deep Neural Net for Detecting Life-Threatening Infant Incidents From Head Images

Tzu-Jui Julius Wang, Jorma Laaksonen, Yi-Ping Liao, Bo-Zong Wu, Shih-Yun Shen

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

49 Downloads (Pure)

Abstract

The notorious incident of sudden infant death syndrome (SIDS) can easily happen to a newborn due to many environmental factors. To prevent such tragic incidents from happening, we propose a multi-task deep learning framework that detects different facial traits and two life-threatening indicators, i.e. which facial parts are occluded or covered, by analyzing the infant head image. Furthermore, we extend and adapt the recently developed models that capture data-dependent uncertainty from noisy observations for our application. The experimental results show significant improvements on YunInfants dataset across most of the tasks over the models that simply adopt the regular cross-entropy losses without addressing the effect of the underlying uncertainties.
Original languageEnglish
Title of host publication2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings
PublisherIEEE
Pages3006-3010
Number of pages5
ISBN (Electronic)9781538662496
DOIs
Publication statusPublished - 2019
MoE publication typeA4 Article in a conference publication
EventIEEE International Conference on Image Processing - Taipei, Taiwan, Republic of China
Duration: 22 Sep 201925 Sep 2019
Conference number: 26

Conference

ConferenceIEEE International Conference on Image Processing
Abbreviated titleICIP
CountryTaiwan, Republic of China
CityTaipei
Period22/09/201925/09/2019

Fingerprint Dive into the research topics of 'A Multi-Task Bayesian Deep Neural Net for Detecting Life-Threatening Infant Incidents From Head Images'. Together they form a unique fingerprint.

  • Projects

    Deep neural networks in scene graph generation for perception of visual multimedia semantics

    Laaksonen, J., Sjöberg, M., Anwer, R., Pehlivan Tort, S. & Wang, T.

    01/01/201831/12/2019

    Project: Academy of Finland: Other research funding

    Equipment

    Science-IT

    Mikko Hakala (Manager)

    School of Science

    Facility/equipment: Facility

  • Cite this

    Wang, T-J. J., Laaksonen, J., Liao, Y-P., Wu, B-Z., & Shen, S-Y. (2019). A Multi-Task Bayesian Deep Neural Net for Detecting Life-Threatening Infant Incidents From Head Images. In 2019 IEEE International Conference on Image Processing, ICIP 2019 - Proceedings (pp. 3006-3010). [8803332] IEEE. https://doi.org/10.1109/ICIP.2019.8803332