Tezislər / Theses


Emotion Recognition from Facial Expressions using Python



Yüklə 17,55 Mb.
Pdf görüntüsü
səhifə174/493
tarix02.10.2023
ölçüsü17,55 Mb.
#151572
1   ...   170   171   172   173   174   175   176   177   ...   493
BHOS Tezisler 2022 17x24sm

Emotion Recognition from Facial Expressions using Python 
Computer Vision is a portion of Artificial Intelligence that deals with visual 
data. With the advent use of machine learning and deep learning models, 


THE 3
rd
 INTERNATIONAL SCIENTIFIC CONFERENCES OF STUDENTS AND YOUNG RESEARCHERS 
dedicated to the 99
th
anniversary of the National Leader of Azerbaijan Heydar Aliyev
183
computer systems today can work with digital images and videos to understand 
and emotionally identify the characteristics of the video’s contents. 
Cognitive Science and Sentiment Analysis 
Artificial Intelligence today has reached heights and lengths that were 
not imagined a few years back. Programs and computer systems can now 
mimic human behavior, reactions, and responses to a high level of accuracy. 
The basic task of any sentiment analysis program is to isolate the polarity 
of the input (text, speech, facial expression, etc.) to understand whether the 
primary sentiment presented is positive, negative, or neutral. Based on this 
initial analysis, programs then often dig deeper to identify emotions like 
enjoyment, happiness, disgust, anger, fear, and surprise. 
Face Emotion Recognizer 
The Face Emotion Recognizer (generally known as the FER) is an open-
source Python library and is used for sentiment analysis of images and 
videos. The project is built on a version that uses a convolution neural 
network with weights mentioned in the HDF5 data of this system’s creation 
model. This can be overridden by using the FER constructor when the model 
is called and initiated. 
1. MTCNN (multi cascade convolutional network) is a parameter of the 
constructor. It is a technique to detect faces. When it is set to ‘True’ the 
MTCNN model is used to detect faces, and when it is set to ‘False’ the 
function uses the default OpenCV Haarcascade classifier. 
2. detect_emotions(): This function is used to classify the detection of 
emotion and it registers the output into six categories, namely, ‘fear’, ‘neutral’, 
‘happy’, ’sad’, ‘anger’, and ‘disgust’. Every emotion is calculated, and the 
output is put on a scale of 0 to 1. 
Conclusion 
The goal of this paper is to present comparative study on different 
techniques of video based facial and emotions recognition using different 
methods. Since from last decade, automatic facial expression and emotions 
recognition plays a significant role in our daily communication and computer 
science, such as human-computer interaction systems, biometrics, and 
security and so on. In recent years, much deep and productive research in 
this area has been carried out. The current methods research problems are 
defined in this paper. For future work, new method designing should be done 
in order to improve robustness, efficiency and accuracy of recognition. 
References 
[1] Landowska, M. Szwoch, W. Szwoch, M.R. Wróbel, A. Kołakowska, Emotion recognition 
and its applications 
[2] Paweł Tarnowski, Marcin Kołodziej, Andrzej Majkowski, Remigiusz J. Rak, Emotion 
recognition using facial expressions, 2017 
[3] Husam Salih, Lalit Kulkarni, Study of Video based Facial Expression and Emotions 
Recognition Mehotds, 2017 



Yüklə 17,55 Mb.

Dostları ilə paylaş:
1   ...   170   171   172   173   174   175   176   177   ...   493




Verilənlər bazası müəlliflik hüququ ilə müdafiə olunur ©azkurs.org 2024
rəhbərliyinə müraciət

gir | qeydiyyatdan keç
    Ana səhifə


yükləyin