Facial Expression Classification Using Convolutional Neural Network and Real Time Application

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

4 Citations (Scopus)

Abstract

Facial expression is an important feature that gives information about a person's psychological situation. People use these expressions while communicating and socializing and they have lots of information related to inner world of an individual. Therefore it is important to understand the meaning of these facial expressions automatically and use this information. This paper presents a method for facial expression classification with grayscale images from Kaggle Face Dataset with a Convolutional Neural Network and a realtime user interface in order to test the performance online. Data augmentation is used to increase the diversity of the samples. Neural network is created with Matlab and user interface is created via App Designer. Different training and fine-tuning techniques are employed in the design. The overall accuracy 61.8% is achieved across seven different facial expression categories with test dataset supplied in Kaggle domain.

Original languageEnglish
Title of host publicationUBMK 2019 - Proceedings, 4th International Conference on Computer Science and Engineering
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages23-27
Number of pages5
ISBN (Electronic)9781728139647
DOIs
Publication statusPublished - Sept 2019
Event4th International Conference on Computer Science and Engineering, UBMK 2019 - Samsun, Turkey
Duration: 11 Sept 201915 Sept 2019

Publication series

NameUBMK 2019 - Proceedings, 4th International Conference on Computer Science and Engineering

Conference

Conference4th International Conference on Computer Science and Engineering, UBMK 2019
Country/TerritoryTurkey
CitySamsun
Period11/09/1915/09/19

Keywords

  • Convolutional Neural Network
  • Data Augmentation
  • Real time testing
  • Training and Fine-Tuning Techniques

Fingerprint

Dive into the research topics of 'Facial Expression Classification Using Convolutional Neural Network and Real Time Application'. Together they form a unique fingerprint.

Cite this