Visual analytics using artificial intelligence : visual events classifier using deep learning

• Main Author: Ong, Kian Kuan
• Other Authors: Yap Kim Hui
• Format: Final Year Project
• Language: English
• Published: 2019
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/77890
• Institution: Nanyang Technological University

As Deep learning emerges from Machine learning to become a leading technology in today’s day and age, there have been many attempts at integrating Deep learning methods into day today applications. Out of these applications, image recognition is the area of interest in which this project aims to elaborate on. Coupled with the explosion of imagery data available worldwide, image recognition has become increasingly popular as a research topic and has continually demonstrated its superiority over traditional Computer Vision and has continually seen rapid development to a point where it has achieved superior performance compared to humans in specific recognition tasks. As the scope of image recognition is non exhaustive, a specific recognition task has to be defined. This project aims to study the feasibility of an events classifier using different state of the art variations of the Convolution Neural network architectures that stems from Deep learning. Through this study, it can be potentially be integrated into a text-based search and retrieval programme in a photo management gallery in storage devices. To support the demonstration of this study, a simple GUI will also be developed for this purpose.