Human emotion recognition in video using subtraction pre-processing

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

Abstract

In this paper, we describe a new image pre-processing method, which can show features or important information clearly. Deep learning methods have grown rapidly in the last ten years and have better performance than the traditional machine learning methods in many domains. Deep learning shows its powerful ability particular in difficult multi-classes classification challenges. Video Facial expression recognition is one of the most popular classification topics and will become essential in robotics and auto-motion fields. The new system presented is a combination of new video pre-processing and Convolutional Neural Network (CNN). The new pre-processing method is proposed because we believe individual emotions are dynamic, which means the change of the face is the key feature. RAVDESS is the video set used, to train and test the neural network. From RAVDESS dataset the video songs without audio are taken for focusing on video frames differences. The chosen video set has six different classes of emotions. Each video presents a sentence in a melodious way. Based on the chosen video set, the new system with a new pre-processing method has been designed and trained. Later, the classification result of the new method has been compared with others in which the same dataset for video emotion recognition was used.
LanguageEnglish
Title of host publicationICMLC '19 Proceedings of the 2019 11th International Conference on Machine Learning and Computing
Place of PublicationNew York
Pages374-379
Number of pages6
DOIs
Publication statusPublished - 22 Feb 2019
Event11th International Conference on Machine Learning and Computing - Zhuhai, China
Duration: 22 Feb 201924 Feb 2019
Conference number: 11
http://www.icmlc.org/

Conference

Conference11th International Conference on Machine Learning and Computing
Abbreviated titleICMLC 2019
CountryChina
CityZhuhai
Period22/02/1924/02/19
Internet address

Fingerprint

Processing
Neural networks
Learning systems
Robotics
Deep learning

Keywords

  • classification
  • video pre-processing
  • images’ difference
  • emotion recognition
  • neural networks

Cite this

He, Z., Jin, T., Basu, A., Soraghan, J., Di Caterina, G., & Petropoulakis, L. (2019). Human emotion recognition in video using subtraction pre-processing. In ICMLC '19 Proceedings of the 2019 11th International Conference on Machine Learning and Computing (pp. 374-379). New York. https://doi.org/10.1145/3318299.3318321
He, Zhihao ; Jin, Tian ; Basu, Amlan ; Soraghan, John ; Di Caterina, Gaetano ; Petropoulakis, Lykourgos. / Human emotion recognition in video using subtraction pre-processing. ICMLC '19 Proceedings of the 2019 11th International Conference on Machine Learning and Computing. New York, 2019. pp. 374-379
@inproceedings{892499597a9e4d64ae548ec4c4779b02,
title = "Human emotion recognition in video using subtraction pre-processing",
abstract = "In this paper, we describe a new image pre-processing method, which can show features or important information clearly. Deep learning methods have grown rapidly in the last ten years and have better performance than the traditional machine learning methods in many domains. Deep learning shows its powerful ability particular in difficult multi-classes classification challenges. Video Facial expression recognition is one of the most popular classification topics and will become essential in robotics and auto-motion fields. The new system presented is a combination of new video pre-processing and Convolutional Neural Network (CNN). The new pre-processing method is proposed because we believe individual emotions are dynamic, which means the change of the face is the key feature. RAVDESS is the video set used, to train and test the neural network. From RAVDESS dataset the video songs without audio are taken for focusing on video frames differences. The chosen video set has six different classes of emotions. Each video presents a sentence in a melodious way. Based on the chosen video set, the new system with a new pre-processing method has been designed and trained. Later, the classification result of the new method has been compared with others in which the same dataset for video emotion recognition was used.",
keywords = "classification, video pre-processing, images’ difference, emotion recognition, neural networks",
author = "Zhihao He and Tian Jin and Amlan Basu and John Soraghan and {Di Caterina}, Gaetano and Lykourgos Petropoulakis",
year = "2019",
month = "2",
day = "22",
doi = "10.1145/3318299.3318321",
language = "English",
isbn = "978-1-4503-6600-7",
pages = "374--379",
booktitle = "ICMLC '19 Proceedings of the 2019 11th International Conference on Machine Learning and Computing",

}

He, Z, Jin, T, Basu, A, Soraghan, J, Di Caterina, G & Petropoulakis, L 2019, Human emotion recognition in video using subtraction pre-processing. in ICMLC '19 Proceedings of the 2019 11th International Conference on Machine Learning and Computing. New York, pp. 374-379, 11th International Conference on Machine Learning and Computing, Zhuhai, China, 22/02/19. https://doi.org/10.1145/3318299.3318321

Human emotion recognition in video using subtraction pre-processing. / He, Zhihao; Jin, Tian; Basu, Amlan; Soraghan, John; Di Caterina, Gaetano; Petropoulakis, Lykourgos.

ICMLC '19 Proceedings of the 2019 11th International Conference on Machine Learning and Computing. New York, 2019. p. 374-379.

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - Human emotion recognition in video using subtraction pre-processing

AU - He, Zhihao

AU - Jin, Tian

AU - Basu, Amlan

AU - Soraghan, John

AU - Di Caterina, Gaetano

AU - Petropoulakis, Lykourgos

PY - 2019/2/22

Y1 - 2019/2/22

N2 - In this paper, we describe a new image pre-processing method, which can show features or important information clearly. Deep learning methods have grown rapidly in the last ten years and have better performance than the traditional machine learning methods in many domains. Deep learning shows its powerful ability particular in difficult multi-classes classification challenges. Video Facial expression recognition is one of the most popular classification topics and will become essential in robotics and auto-motion fields. The new system presented is a combination of new video pre-processing and Convolutional Neural Network (CNN). The new pre-processing method is proposed because we believe individual emotions are dynamic, which means the change of the face is the key feature. RAVDESS is the video set used, to train and test the neural network. From RAVDESS dataset the video songs without audio are taken for focusing on video frames differences. The chosen video set has six different classes of emotions. Each video presents a sentence in a melodious way. Based on the chosen video set, the new system with a new pre-processing method has been designed and trained. Later, the classification result of the new method has been compared with others in which the same dataset for video emotion recognition was used.

AB - In this paper, we describe a new image pre-processing method, which can show features or important information clearly. Deep learning methods have grown rapidly in the last ten years and have better performance than the traditional machine learning methods in many domains. Deep learning shows its powerful ability particular in difficult multi-classes classification challenges. Video Facial expression recognition is one of the most popular classification topics and will become essential in robotics and auto-motion fields. The new system presented is a combination of new video pre-processing and Convolutional Neural Network (CNN). The new pre-processing method is proposed because we believe individual emotions are dynamic, which means the change of the face is the key feature. RAVDESS is the video set used, to train and test the neural network. From RAVDESS dataset the video songs without audio are taken for focusing on video frames differences. The chosen video set has six different classes of emotions. Each video presents a sentence in a melodious way. Based on the chosen video set, the new system with a new pre-processing method has been designed and trained. Later, the classification result of the new method has been compared with others in which the same dataset for video emotion recognition was used.

KW - classification

KW - video pre-processing

KW - images’ difference

KW - emotion recognition

KW - neural networks

UR - http://www.icmlc.org/

U2 - 10.1145/3318299.3318321

DO - 10.1145/3318299.3318321

M3 - Conference contribution book

SN - 978-1-4503-6600-7

SP - 374

EP - 379

BT - ICMLC '19 Proceedings of the 2019 11th International Conference on Machine Learning and Computing

CY - New York

ER -

He Z, Jin T, Basu A, Soraghan J, Di Caterina G, Petropoulakis L. Human emotion recognition in video using subtraction pre-processing. In ICMLC '19 Proceedings of the 2019 11th International Conference on Machine Learning and Computing. New York. 2019. p. 374-379 https://doi.org/10.1145/3318299.3318321