Picture for Marthie Grobler

Marthie Grobler

Resurrecting Trust in Facial Recognition: Mitigating Backdoor Attacks in Face Recognition to Prevent Potential Privacy Breaches

Add code
Feb 18, 2022
Figure 1 for Resurrecting Trust in Facial Recognition: Mitigating Backdoor Attacks in Face Recognition to Prevent Potential Privacy Breaches
Figure 2 for Resurrecting Trust in Facial Recognition: Mitigating Backdoor Attacks in Face Recognition to Prevent Potential Privacy Breaches
Figure 3 for Resurrecting Trust in Facial Recognition: Mitigating Backdoor Attacks in Face Recognition to Prevent Potential Privacy Breaches
Figure 4 for Resurrecting Trust in Facial Recognition: Mitigating Backdoor Attacks in Face Recognition to Prevent Potential Privacy Breaches
Viaarxiv icon

Robust Training Using Natural Transformation

Add code
May 10, 2021
Figure 1 for Robust Training Using Natural Transformation
Figure 2 for Robust Training Using Natural Transformation
Figure 3 for Robust Training Using Natural Transformation
Figure 4 for Robust Training Using Natural Transformation
Viaarxiv icon

OCTOPUS: Overcoming Performance andPrivatization Bottlenecks in Distributed Learning

Add code
May 03, 2021
Figure 1 for OCTOPUS: Overcoming Performance andPrivatization Bottlenecks in Distributed Learning
Figure 2 for OCTOPUS: Overcoming Performance andPrivatization Bottlenecks in Distributed Learning
Figure 3 for OCTOPUS: Overcoming Performance andPrivatization Bottlenecks in Distributed Learning
Figure 4 for OCTOPUS: Overcoming Performance andPrivatization Bottlenecks in Distributed Learning
Viaarxiv icon

Adversarial Defense by Latent Style Transformations

Add code
Jun 17, 2020
Figure 1 for Adversarial Defense by Latent Style Transformations
Figure 2 for Adversarial Defense by Latent Style Transformations
Figure 3 for Adversarial Defense by Latent Style Transformations
Figure 4 for Adversarial Defense by Latent Style Transformations
Viaarxiv icon

Defending Adversarial Attacks via Semantic Feature Manipulation

Add code
Feb 03, 2020
Figure 1 for Defending Adversarial Attacks via Semantic Feature Manipulation
Figure 2 for Defending Adversarial Attacks via Semantic Feature Manipulation
Figure 3 for Defending Adversarial Attacks via Semantic Feature Manipulation
Figure 4 for Defending Adversarial Attacks via Semantic Feature Manipulation
Viaarxiv icon

OIAD: One-for-all Image Anomaly Detection with Disentanglement Learning

Add code
Jan 18, 2020
Figure 1 for OIAD: One-for-all Image Anomaly Detection with Disentanglement Learning
Figure 2 for OIAD: One-for-all Image Anomaly Detection with Disentanglement Learning
Figure 3 for OIAD: One-for-all Image Anomaly Detection with Disentanglement Learning
Figure 4 for OIAD: One-for-all Image Anomaly Detection with Disentanglement Learning
Viaarxiv icon

Backdoor Attacks against Transfer Learning with Pre-trained Deep Learning Models

Add code
Jan 10, 2020
Figure 1 for Backdoor Attacks against Transfer Learning with Pre-trained Deep Learning Models
Figure 2 for Backdoor Attacks against Transfer Learning with Pre-trained Deep Learning Models
Figure 3 for Backdoor Attacks against Transfer Learning with Pre-trained Deep Learning Models
Figure 4 for Backdoor Attacks against Transfer Learning with Pre-trained Deep Learning Models
Viaarxiv icon

Generating Semantic Adversarial Examples via Feature Manipulation

Add code
Jan 06, 2020
Figure 1 for Generating Semantic Adversarial Examples via Feature Manipulation
Figure 2 for Generating Semantic Adversarial Examples via Feature Manipulation
Figure 3 for Generating Semantic Adversarial Examples via Feature Manipulation
Figure 4 for Generating Semantic Adversarial Examples via Feature Manipulation
Viaarxiv icon

Security and Performance Considerations in ROS 2: A Balancing Act

Add code
Sep 24, 2018
Figure 1 for Security and Performance Considerations in ROS 2: A Balancing Act
Figure 2 for Security and Performance Considerations in ROS 2: A Balancing Act
Figure 3 for Security and Performance Considerations in ROS 2: A Balancing Act
Figure 4 for Security and Performance Considerations in ROS 2: A Balancing Act
Viaarxiv icon