Adversarial Attacks on Monocular Pose Estimation

Advances in deep learning have resulted in steady progress in computer vision with improved accuracy on tasks such as object detection and semantic segmentation. Nevertheless, deep neural networks are vulnerable to adversarial attacks, thus presenting a challenge in reliable deployment. Two of the prominent tasks in 3D scene-understanding for robotics and advanced driver assistance systems are monocular depth and pose estimation, often learned together in an unsupervised manner. While studies evaluating the impact of adversarial attacks on monocular depth estimation exist, a systematic demonstration and analysis of adversarial perturbations against pose estimation are lacking. We show how additive imperceptible perturbations can not only change predictions to increase the trajectory drift but also catastrophically alter its geometry. We also study the relation between adversarial perturbations targeting monocular depth and pose estimation networks, as well as the transferability of perturbations to other networks with different architectures and losses. Our experiments show how the generated perturbations lead to notable errors in relative rotation and translation predictions and elucidate vulnerabilities of the networks. 1 1 Code can be found at https://github.com/NeurAI-Lab/mono-pose-attack.

Authors

• Hemang Chawla
• Arnav Varma
• Elahe Arani
• Bahram Zonooz

Keywords

• Deep learning
• Training
• Three-dimensional displays
• Systematics
• Perturbation methods
• Pose estimation
• Predictive models
• Adversarial Attacks
• Monocular Pose Estimation
• Deep Neural Network
• Prediction Error
• Object Detection
• Semantic Segmentation
• Imperceptible
• Estimation Network
• Depth Estimation
• Unsupervised Manner
• Advances In Deep Learning
• Advanced Driver Assistance Systems
• Adversarial Perturbations
• Monocular Depth Estimation
• Root Mean Square Error
• Input Image
• Image Pairs
• Image Generation
• Clear Image
• Relative Pose
• Pose Prediction
• Camera Pose
• Network Depth
• Rotation Error
• Geometric Loss
• Training Loss
• Projected Gradient Descent
• Self-driving
• Translation Error