CloudFort: Enhancing Robustness of 3D Point Cloud Classification Against Backdoor Attacks via Spatial Partitioning and Ensemble Prediction

arXiv:2404.14042v1 Announce Type: new
Abstract: The increasing adoption of 3D point cloud data in various applications, such as autonomous vehicles, robotics, and virtual reality, has brought about significant advancements in object recognition and scene understanding. However, this progress is accompanied by new security challenges, particularly in the form of backdoor attacks. These attacks involve inserting malicious information into the training data of machine learning models, potentially compromising the model's behavior. In this paper, we propose CloudFort, a novel defense mechanism …

abstract adoption applications arxiv attacks autonomous autonomous vehicles backdoor classification cloud cloud data cs.cv data ensemble however object partitioning prediction progress reality recognition robotics robustness security spatial type understanding vehicles via virtual virtual reality