arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.08344cs.CV
收藏

基于卷积神经网络的伪造人脸视频检测

Detecting Forged Facial Videos using convolutional neural network

University of South Florida(南佛罗里达大学)

Neilesh Sambhu, Shaun Canavan

摘要:

在本文中,我们提出了检测伪造的视频,人脸,在线视频。为了促进这种检测,我们建议使用更小(更少的参数学习)的卷积神经网络(CNN),用于数据驱动的伪造视频检测方法。为了验证我们的方法,我们调查了FaceForensics公共数据集,详细介绍了基于帧和基于视频的结果。所提出的方法被证明优于当前最先进的。我们还进行了消融研究,分析批量大小,过滤器数量和网络层数对检测伪造视频的准确性的影响。

英文摘要:

In this paper, we propose to detect forged videos, of faces, in online videos. To facilitate this detection, we propose to use smaller (fewer parameters to learn) convolutional neural networks (CNN), for a data-driven approach to forged video detection. To validate our approach, we investigate the FaceForensics public dataset detailing both frame-based and video-based results. The proposed method is shown to outperform current state of the art. We also perform an ablation study, analyzing the impact of batch size, number of filters, and number of network layers on the accuracy of detecting forged videos.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑