Facial Sentiment Analysis

Project Title

Facial Sentiment Analysis

Project Area of Specialization Artificial IntelligenceProject Summary

Facial Expression Recognition is the process of mapping facial expressions on human faces. More specifically, this technology is a sentiment analysis tool and can automatically detect the seven basic or universal expressions: happiness, sadness, anger, surprise, fear, neutral disgust, in addition to that it can detect drowsiness for one or a group of people present in the image.
FER is a challenging task, the problem of expression recognition for a group of people has been less extensively studied. Expression Recognition is a challenging problem due to obstructions like head and body pose variations, occlusions, variable lighting conditions, variance of actors, varied indoor and outdoor settings and image quality.

Project Objectives

• This project will carry out performance analysis of the different methods and models, to train a good machine learning for both emotion recognition and drowsiness detection model and create a pipeline for integration.
• In this project, we will develop a mobile application, that allows users to upload an image for expression recognition from their gallery or taken immediately from the camera of their smartphone/tablet.
• This application will have a meeting mode to identify the perceived expression of each person in the meeting separately and the overall sentiments as a whole, and eventually create a report.
• This application will also have a music suggestion system to suggest some music by taking an image to analyze the sentiments of people in it to play it accordingly.

Project Implementation Method

First of all, we will do software requirement engineering which includes a detailed project plan with work breakdown structure (WBS), requirements specifications, business requirements, user requirements, functional requirements, non-functional requirements, hardware requirements, Use
case diagram(s) etc.

As this project uses machine learning techniques and the first for that is gathering data, for that we will use an online data repository kaggle but the data available is not enough so we will take the initiative to create our own additional data by taking pictures in my campus, The next step is to preprocess those images using python as a programming language.

Now that the data would have been preprocessed, we will then train multiple machine learning models using multiple python libraries and then carry out a performance analysis of the different methods and models to choose the best one to proceed with.

After the model would have been trained and working properly we will create an application utilizing this model's functionality that would provide a meeting mode to identify the perceived expression of each person in the meeting separately and the overall sentiments as a whole, and eventually create a report, in addition to that, we would integrate a music suggestion system to suggest some music by taking an image to analyze the sentiments of people in it to play it accordingly.

Benefits of the Project

There are many benefits to facial expressions recognition:

• Identifying the sentiments of students in a classroom to know whether they are interested in the lecture or not, This module is also included in our project.
• By using this technique smart glasses can help autistic/blind children interpret the feelings of people around them. 
• The healthcare industry sure can take advantage of facial expression recognition, they can use it to know if a patient needs medicine or for physicians to know who to prioritize in seeing first.
• Companies can also use expression recognition to determine their business outcomes in terms of the audience’s facial responses.
• The system can help analyse the mood and recommend things accordingly, as in our case the system analyzes the mood of the user by facial expressions to play the music accordingly.

Technical Details of Final Deliverable

The final deliverable includes all types of Software requirements engineering documents i.e.:

• A detailed project plan with a work breakdown structure (WBS)
• Requirements specifications(business requirements user requirements, functional requirements, non-functional requirements, hardware requirements)
• Use case diagram(s)
• Prototypes (Mockups)
• System Architecture
• Detailed System Design (ER/EER Diagram, UCD, Class Diagram, Sequence Diagram, Collaboration Diagram)
• Use cases description,
• Test plans, Test cases

A Mobile and Web application using technologies like React, Python, Bootstrap, Django, etc. that have functionalities:

• A music player that uses facial expressions to analyze which music should be played based on the sentiments smartly.
• A meeting analyzer that accesses a video feed to analyze the sentiments of people present in it and create a report based on it, this meeting can be a classroom, an interrogation room, a healthcare facility, or a business meeting.

Final Deliverable of the Project Software SystemCore Industry EducationOther Industries Media , Health , Security Core Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Quality EducationRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	66000
Stationery	Miscellaneous	10	100	1000
Commuting	Miscellaneous	4	2000	8000
SSD	Equipment	4	5500	22000
Graphic Card	Equipment	1	35000	35000