Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/138764
Title: Visual analytics using artificial intelligence : visual events classifier using deep learning
Authors: Koh, Josephine Si Ting
Keywords: Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
Engineering::Electrical and electronic engineering
Issue Date: 2020
Publisher: Nanyang Technological University
Project: A3289-191
Abstract: With the evolution of camera features on mobile phones, users often have a difficult time rummaging for specific photos in their huge gallery collection. Thus, this project aims to develop a text-based search application based on deep learning techniques. As Deep Learning are widely used in many different applications, this project narrows down to explore the classification of Singapore related events. This project would study the different methods that are being used to develop an image classifier. With the suitable method chosen, images will be gathered manually to construct the dataset required. Using transfer learning techniques, different models will be studied and tested as a comparison. The best performing model would be used in the text-based image search application. The last part of the project, a GUI application would be created and integrated using multiple classifiers namely events classifier, landmark classifier and group emotion recognition. The application would enable users to search and retrieve images using the keywords that they enter. Lastly, several discussions will be done on how the current model and dataset can be further improved to develop a more robust classifier.
URI: https://hdl.handle.net/10356/138764
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
FYP Final Report_Josephine.pdf
  Restricted Access
20.16 MBAdobe PDFView/Open

Page view(s)

248
Updated on Feb 7, 2023

Download(s) 50

24
Updated on Feb 7, 2023

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.