文章基本信息

标题：Detection and Mitigating Facial Recognition Classifiers against Adversarial Attacks
本地全文：下载
作者：S.Chinnadurai ; R.Hariprasath ; V.Aravinth 等
期刊名称：International Journal of Advances in Engineering and Management
电子版ISSN：2395-5252
出版年度：2021
卷号：3
期号：5
页码：913-918
DOI：10.35629/5252-0305785791
语种：English
出版社：IJAEM JOURNAL
摘要：The aim of this project is to present a unified framework for human action and activity recognition.Black box models use the classification accuracy of the performance metric for validating their defense. This is a problem for biometrics verification applications where the incoming image is not cannot compute the accuracy of the classifier.Attacks is a critical step of deep learning solutions for biometrics verification. It proposes a novel framework for defending Black box systems from adversarial attacks using an ensemble of iterative adversarial image purifiers performance is using Bayesian uncertainties. This is used to estimate an intraprediction mode from a prediction unit and to reduce encoding time significantly by avoiding the intensive Rate-Distortion optimization of a number of intra-prediction modes. The proposed technique is High Efficiency Video Coding Test Model (HM) video coding standard and Joint Exploration Model (JEM) reference software, by integrating the random forest trained off-line into the codecs.To summarize a set of events and to search for particular events because they contain various pieces of context information.