Toward Robust 3D Perception for Autonomous Vehicles: A Review of Adversarial Attacks and Countermeasures
At present the perception system of autonomous vehicles is grounded on 3D vision
technologies along with deep learning to process depth information. Although deep learning …
technologies along with deep learning to process depth information. Although deep learning …
Poly-CAM: High resolution class activation map for convolutional neural networks
The demand for explainable AI continues to rise alongside advancements in deep learning
technology. Existing methods such as convolutional neural networks often struggle to …
technology. Existing methods such as convolutional neural networks often struggle to …
Face3DAdv: Exploiting Robust Adversarial 3D Patches on Physical Face Recognition
Recent research has elucidated the susceptibility of face recognition models to physical
adversarial patches, thus provoking security concerns about the deployed face recognition …
adversarial patches, thus provoking security concerns about the deployed face recognition …
3D Visual Grounding-Audio: 3D scene object detection based on audio
C Zhang, Z Cai, X Chen, F Da, S Gai - Neurocomputing, 2024 - Elsevier
Abstract 3D Visual Grounding (3DVG) is a prevalent multi-modal information fusion task
capable of accurately localizing target objects referenced in natural language descriptions …
capable of accurately localizing target objects referenced in natural language descriptions …
Prototypical Transformer as Unified Motion Learners
In this work, we introduce the Prototypical Transformer (ProtoFormer), a general and unified
framework that approaches various motion tasks from a prototype perspective. ProtoFormer …
framework that approaches various motion tasks from a prototype perspective. ProtoFormer …
Self-supervised Adversarial Training of Monocular Depth Estimation against Physical-World Attacks
Monocular Depth Estimation (MDE) plays a vital role in applications such as autonomous
driving. However, various attacks target MDE models, with physical attacks posing …
driving. However, various attacks target MDE models, with physical attacks posing …
ControlLoc: Physical-World Hijacking Attack on Visual Perception in Autonomous Driving
Recent research in adversarial machine learning has focused on visual perception in
Autonomous Driving (AD) and has shown that printed adversarial patches can attack object …
Autonomous Driving (AD) and has shown that printed adversarial patches can attack object …
Dyna-MSDepth: multi-scale self-supervised monocular depth estimation network for visual SLAM in dynamic scenes
J Yao, Y Li, J Li - Machine Vision and Applications, 2024 - Springer
Abstract Monocular Simultaneous Localization And Mapping (SLAM) suffers from scale drift,
leading to tracking failure due to scale ambiguity. Deep learning has significantly advanced …
leading to tracking failure due to scale ambiguity. Deep learning has significantly advanced …
[PDF][PDF] Trustworthy and Robust Machine Learning for Multimedia: Challenges and Perspectives
Multimedia applications for machine learning models are characterized by the fusion of
multiple modalities of data. In this work, we highlight the trust and robustness challenges of …
multiple modalities of data. In this work, we highlight the trust and robustness challenges of …