This paper aims to improve privacy-preserving visual recognition, an increasingly demanded feature in smart camera applications, by formulating a unique adversarial training framework. The proposed framework explicitly learns a degradation transform for the original video inputs, in order to optimize the trade-off between target task performance and the associated privacy budgets on the degraded video. A notable challenge is that the privacy budget, often defined and measured in task-driven contexts, cannot be reliably indicated using any single model performance, because a strong protection of privacy has to sustain against any possible model that tries to hack privacy information. Such an uncommon situation has motivated us to propose two strategies, i.e., budget model restarting and ensemble, to enhance the generalization of the learned degradation on protecting privacy against unseen hacker models. Novel training strategies, evaluation protocols, and result visualization methods have been designed accordingly. Two experiments on privacypreserving action recognition, with privacy budgets defined in various ways, manifest the compelling effectiveness of the proposed framework in simultaneously maintaining high target task (action recognition) performance while suppressing the privacy breach risk. The code is available at https://github.com/wuzhenyusjtu/Privacy-AdversarialLearning
Object detection from images captured by Unmanned Aerial Vehicles (UAVs) is becoming increasingly useful. Despite the great success of the generic object detection methods trained on ground-to-ground images, a huge performance drop is observed when they are directly applied to images captured by UAVs. The unsatisfactory performance is owing to many UAV-specific nuisances, such as varying flying altitudes, adverse weather conditions, dynamically changing viewing angles, etc. Those nuisances constitute a large number of fine-grained domains, across which the detection model has to stay robust. Fortunately, UAVs will record meta-data that depict those varying attributes, which are either freely available along with the UAV images, or can be easily obtained. We propose to utilize those free meta-data in conjunction with associated UAV images to learn domain-robust features via an adversarial training framework dubbed Nuisance Disentangled Feature Transform (NDFT), for the specific challenging problem of object detection in UAV images, achieving a substantial gain in robustness to those nuisances. We demonstrate the effectiveness of our proposed algorithm, by showing state-ofthe-art performance (single model) on two existing UAVbased object detection benchmarks. The code is available at https:// github.com/ TAMU-VITA/ UAV-NDFT.
Account compromisation is a serious threat to users of Online Social Networks (OSNs). While relentless spammers exploit the established trust relationships between account owners and their friends to efficiently spread malicious spam, timely detection of compromised accounts are quite challenging due to the well established trust relationship between the service providers, account owners, and their friends. In this paper, we study the social behaviors of OSN users, i.e. their usage of OSN services, and the application of which in detecting compromised accounts. In particular, we propose a set of social behavioral features that can effectively characterize the user social activities on OSNs. We validate the efficacy of these behavioral features by collecting and analyzing real user clickstreams to an OSN website. Based on our measurement study, we devise individual user's social behavioral profile by combining its respective behavioral feature metrics. A social behavioral profile accurately reflects a user's OSN activity patterns. While an authentic owner conforms to its account's social behavioral profile involuntarily, it is hard and costly for impostors to feign. We evaluate the capability of the social behavioral profiles in distinguishing different OSN users, and our experimental results show the social behavioral profiles can accurately differentiate individual OSN users and detect compromised accounts.1556-6013 (c)
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.