Skip to content

holman57/AI_Emotion_Detection

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AI Emotion Recognition

Spring 2020 AI project

https://docs.google.com/document/d/1PjkWR5__LrFzUJAlzkBKJEJTB5ch_Uyd/edit

Abstract

Research has indicated that human emotion can be detected via facial expressions and speech, and identifying these emotions may be a quality innate to most humans.

Distinctly classifying these emotions can be difficult, and prime human emotions have been even more difficult to identify. Machine learning models allow us to not only solve prediction problems, but see how different models solve these problems in a specific domain.

By identifying underlying patterns and using dimensionality reduction via machine learning, we may be able to better understand the distinctions and similarities between human emotions and how we perceive them in others.

We hypothesize that we may be able to identify a set of prime emotions that are recognizable by machine learning models.. This will be accomplished by gathering large datasets of facial expressions and speech clips displaying different levels of emotions.

A model that can accurately detect human emotions in individuals will help social services such as psychiatry, rehabilitation and counselling, by helping bridge any gaps in empathy between patient and caregiver.

Introduction

Introduction What is the problem you are addressing?

  • Humans may be able to identify emotions in others fairly easily. However, quantifying these emotions is more difficult, and methods and classifications of recognizable human emotions are debated.

What very specific claim do you hope to make when your work is complete?

  • We will show that our method of recognizing emotions from images of faces will improve the F1-measure of the classifier by a relative 33% compared to choosing the most frequent class when applied to images of unseen people’s faces.

Who will care about your system?

  • Caregivers that are required to work with patients directly in situations where the patients current emotions are important to understand, such as counsellors, psychiatrists, therapists, etc.

Why do they care?

  • It will aid them in understanding their patients emotional state.

Dataset

About

Spring 2020 AI project on emotion detection

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published