Backdoor Attack on gaze estimation

Title

Backdoor Attack on gaze estimation

 Author

Reda, Yuji (TU Delft Electrical Engineering, Mathematics and Computer Science)

Contributor

Du, L. (mentor)
Lan, G. (graduation committee)
Zhang, X. (graduation committee)

Degree granting institution

Delft University of Technology

 Programme

Computer Science and Engineering

 Project

CSE3000 Research Project

 Date

2023-06-25

 Abstract

Badnets are a type of backdoor attack that aims at manipulating the behavior of Convolutional Neural Networks. The training is modified such that when certain triggers appear in the inputs the CNN is going to behave accordingly. In this paper, we apply this type of backdoor attack to a regression task on gaze estimation. We examine different triggers to discover which of them lead to better performance and thus infer which trigger aspects one can take the most advantage from. It turns out that placing frames around and drawing multiple lines across the images are the most effective for the training of Badnets.

 Subject

Gaze estimation
Convolutional Neural Network
Backdoor Attacks
Badnets

 To reference this document use:

http://resolver.tudelft.nl/uuid:80cfdc30-f335-41a8-8665-83f92265edc0

 Part of collection

Student theses

 Document type

bachelor thesis

 Rights

© 2023 Yuji Reda
Files
PDF CSE3000_report.pdf 5.33 MB
Close viewer