arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2512.18809cs.CVcs.AIcs.MM
收藏

FedVideoMAE:高效的隐私保护联邦视频审核框架

FedVideoMAE: Efficient Federated Video Moderation with Differential Privacy and Secure Aggregation

Macquarie University(麦考瑞大学);The University of Sydney(悉尼大学)

Ziyuan Tao, Chuanzhi Xu, Sandaru Jayawardana, Adnan Mahmood, Wei Bao, Kanchana Thilakarathna, Teng Joon Lim

AI总结:

FedVideoMAE结合自监督VideoMAE表示、LoRA参数高效适应、客户端DP-SGD和服务器端安全聚合,通过仅更新550万参数,降低通信成本28.3倍,实现隐私保护下的视频审核。

摘要:

短视频审核日益需要保护用户隐私且不支付云计算推理全部带宽和延迟成本的学习管道。我们提出了FedVideoMAE,一种用于视频暴力检测的设备端联邦框架,结合自监督VideoMAE表示、基于LoRA的参数高效适应、客户端侧DP-SGD和服务器侧安全聚合。通过仅更新550万个参数(约1560万个骨干参数的3.5%),FedVideoMAE相对于全模型联邦更新减少了28.3倍的通信成本,同时在整个训练过程中保持原始视频在设备上。在RWF-2000数据集上,使用40个客户端时,该方法在无隐私保护时达到77.25%的准确率,在强差分隐私下达到65-66%。我们进一步表明,这种隐私差距与针对小数据、参数高效联邦环境定制的有效SNR分析一致,表明在我们的设置中大约有8.5-12倍的DP噪声放大。为了更清晰地定位这些结果,我们还对比了存档的全模型联邦基线,并总结了在RLVS和二元UCF-Crime上的辅助转移行为。总的来说,这些发现将FedVideoMAE定位为在边缘设备上进行隐私保护视频审核的实用操作点。我们的代码可在:https://github.com/zyt-599/FedVideoMAE找到。

英文摘要:

Short-form video moderation is increasingly pushed toward edge and privacy-sensitive settings, where users may intend videos for a limited audience, such as friends or private groups, but sending raw clips to a central server can broaden exposure, consume bandwidth, and add moderation latency. Federated learning can keep videos on device, but unprotected model updates may still leak information, and full-video backbones are expensive to communicate. We present FedVideoMAE, a privacy-preserving federated framework for violence detection that adapts a frozen VideoMAE backbone with lightweight LoRA and prompt parameters. Each training round combines self-supervised masked video reconstruction with client-side differential privacy and pairwise masked aggregation (SA) of adapter updates. Violence labels are held out from federation and used only for downstream evaluation, separating private representation learning from supervised assessment. On RWF-2000, exchanging 5,518,848 trainable parameters instead of the 156,371,328-parameter instantiated pretraining state gives a 28.3x model-state payload ratio. FedVideoMAE reaches 77.25% test accuracy without DP or SA, while accuracy under DP+SA remains in the 65.25-66.00% range. Transfer experiments on RLVS and binary UCF-Crime show similar behavior. These results characterize the privacy-utility trade-off for edge video moderation. Code is available at: https://github.com/zyt-599/FedVideoMAE

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。

↑