GenLie通过全局监督提升微表情检测能力，有效抑制身份噪声干扰。
GenLie: A Global-Enhanced Lie Detection Network under Sparsity and Semantic Interference
- 局部捕捉细微说谎信号，全局监督抑制身份相关噪声
- 在三个公开数据集上均超越现有最优方法，高/低风险场景皆适用
- 适合关注视频说谎识别与鲁棒特征学习的研究者
基于视频的说谎检测旨在从视觉线索中识别欺骗行为。尽管近期取得进展，其核心挑战在于学习稀疏且具有区分性的表征。欺骗信号通常细微且短暂，易被冗余信息掩盖，个体差异和上下文变化引入强烈的身份相关噪声。为此，我们提出GenLie——一种全局增强的说谎检测网络，在全局监督下进行局部特征建模。具体而言，局部层面捕捉稀疏而微妙的欺骗线索，全局监督与优化则通过抑制身份相关噪声，确保表征的鲁棒性与区分性。在三个公开数据集上的实验表明，覆盖高风险与低风险场景，GenLie始终优于现有最优方法。源代码已公开于 https://github.com/AliasDictusZ1/GenLie。
原文摘要 · Abstract (English)
Video-based lie detection aims to identify deceptive behaviors from visual cues. Despite recent progress, its core challenge lies in learning sparse yet discriminative representations. Deceptive signals are typically subtle and short-lived, easily overwhelmed by redundant information, while individual and contextual variations introduce strong identity-related noise. To address this issue, we propose GenLie, a Global-Enhanced Lie Detection Network that performs local feature modeling under global supervision. Specifically, sparse and subtle deceptive cues are captured at the local level, while global supervision and optimization ensure robust and discriminative representations by suppressing identity-related noise. Experiments on three public datasets, covering both high- and low-stakes scenarios, show that GenLie consistently outperforms state-of-the-art methods. Source code is available at https://github.com/AliasDictusZ1/GenLie.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。