Deep learning-enabled invisibility cloak design

Many people have been fascinated with the topic of invisibility since a long time ago and there have been many invisibility cloaks created throughout the years. In this Thesis, a new type of invisibility cloak design is proposed to lessen the effort in creating invisibility cloaks. The proposed invi...

Full description

Saved in:
Bibliographic Details
Main Author: Lim, Cheryl Jing Xuan
Other Authors: Luo Yu
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2023
Subjects:
Online Access:https://hdl.handle.net/10356/167117
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Many people have been fascinated with the topic of invisibility since a long time ago and there have been many invisibility cloaks created throughout the years. In this Thesis, a new type of invisibility cloak design is proposed to lessen the effort in creating invisibility cloaks. The proposed invisibility cloak design is entirely made using deep learning models, namely the You only look once (YOLO) detection model and the Guided Language to Image Diffusion for Generation and Editing (GLIDE) generative model. The two models are linked by the creation of a mask, which results in an algorithm that can keep an object out of sight in an image, hence disguising the object.