2. National Education Society®
S.R NAGAPPA SHETTY MEMORIAL NATIONAL
COLLEGE OF APPLIED SCIENCE,SHIMOGA.
DEPARTMENT OF BCA
SEMINAR ON
REAL-TIME AGE GENDER EMOTION AND RACE RECOGNITION
USING MACHINE LEARNING
TEAM MEMBERS
• ARUN N SHET (BC200520) • ROBIN M P (BC200598)
• VENKATESHA G (BC200638) • ANKITH A S (BC200513)
GUIDED BY,
Ms SHWETHAY K ,
ASST PROF , Dept Of BCA,
SRNMNC,SHIVAMOGGA.
3. ABSTRACT:
o In this we describe a methodology and an algorithm to estimate the
Real-time Age, Gender, Race and Emotion of a human by analysing of
face images on the webcam.
o Here we discuss the CNN based architecture to design a Real-time model.
o Emotion ,Gender Race and Age detection of facial images in the webcam
play an important role in many applications like forensic, security control,
data analysis, video observation and human computer interactions.
4. CONTENTS:
• INTRODUCTION
• SCOPE OF THE PROJECT
• EXISTING SYSTEM
• PROPOSED SYSTEM
• REQUIREMENTS
• SYSTEM DESIGN
• CURRENT PROBLEM IN PROJECT
• AREA OF IMPROVEMENT
• FUTURE SCOPE OF PROJECT
• CONCLUSION
• BIBLIOGRAPHY
5. INTRODUCTION :
o Facial attribute recognition, including age, gender , race and emotion, has been a
topic of interest among computer vision researchers for over a decade.
o One of the key reasons is the numerous applications of this challenging problem
which range from security control, to person identification, to human-computer
interaction.
o Due to the release of large labeled datasets, as well as the advances made in the
design of convolutional neural networks, error rates have dropped significantly. In
many cases, these systems are able to outperform humans.
o However, this still remains a difficult problem and existing commercial systems
fall short when dealing with real world scenarios.
o In this work, we present an end-to-end system capable of estimating facial
attributes including age, gender,race and emotion with low error rates.
6. SCOPE OF PROJECT:
The scope of age, gender, race, and emotion detection using machine learning is
quite broad and can be applied in various domains and applications.
Here are some examples of the scope in each area:
Gender Detection:
Gender-based marketing and advertising: Machine learning models can analyze
user data and predict gender, enabling targeted advertising campaigns.
Race/Ethnicity Detection:
Demographic analysis: Machine learning models can classify individuals into
different racial or ethnic groups based on visual cues, which can be used for
demographic research or social analysis.
7. EXISTING SYSTEM:
Older systems may have more limited capabilities and may not have the same level
of accuracy or robustness as more recent advancements in age, gender, race, and
emotion detection using machine learning.
8. PROPOSED SYSTEM :
The primary objective of the proposed system is to recognize the Gender
and Age range with Emotion from the human face images utilizing the set
of facial features in Real-time applications.
Feature extraction from face images is an important part of this method .
10. Web Camera: This component captures the video feed that will be used by the system.
Pre-processing: This component is responsible for cleaning and pre-processing the video
feed to make it ready for analysis.
Feature Extraction: This component is responsible for extracting the age, gender, emotion,
and race features from the pre-processed video feed.
Machine Learning Model: This component is responsible for analyzing the extracted
features to detect the age,gender, emotion, and race of a person from the video feed.
User Interface: This component is responsible for displaying the detected attributes in real-
time and providing a report on customer satisfaction based on the detected attributes.
11. Usually age, gender ,emotion and race are detected individually in
traditional methods.
13. FUTURE SCOPE OF THE PROJECT:
Storing the details on the database.
Developing graphical user interface for the proposed system.
14.
15. Busso,Carlos,e.a.: Analysis of emotion recognition using facial expressions, speech and
multimodal information. In: Proceedings of the 6th international conference on Multimodal
interfaces. (2004).
Levi, G., Hassner., T.: Emotion recognition in the wild via convolutional neural networks and
mapped binary patterns. In: Proceedings of the 2015 ACM on International Conference on
Multimodal Interaction. ACM. (2015).
Pang, L., Ngo., C.W.: Mutlimodal learning with deep boltzmann machine for emotion
prediction in user generated videos. In: Proceedings of the 2015 ACM on International
Conference on Multimodal Retrieval. ACM. (2015).
Wang, X., Guo, R., Kambhamettu, C.: Deeply-learned feature for age estimation.In:
WACV. (2015).
Gallagher, A.C., Chen., T.: Understanding images of groupsof people. In: CVPR.(2009).
BIBLIOGRAPHY: