Visual questioning and answering

With the rising trend of artificial intelligence and machine learning, more and more intelligent tasks thought to be previously impossible are now feasible and capable to be implemented and automated by machines. This project will look into the implementation and automation of one such task, a Visua...

Full description

Saved in:
Bibliographic Details
Main Author: Ong, Zavier Jian Le
Other Authors: Hanwang Zhang
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2024
Subjects:
Online Access:https://hdl.handle.net/10356/175465
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:With the rising trend of artificial intelligence and machine learning, more and more intelligent tasks thought to be previously impossible are now feasible and capable to be implemented and automated by machines. This project will look into the implementation and automation of one such task, a Visual Question and Answering system. The system accepts a query in the form of a natural language and an input image, and outputs a response similarly in the form of a natural language. This system demonstrates the challenges of computer vision and natural language processing tied to work together harmoniously to produce an accurate output. Thus, this project will showcase an architectural design of this system as well as the implementation method used to demo the aforementioned.