CT-ISG: Trustworthy Media: Blind Passive Media Forensics for Content Integrity Verification

提案：0716203 Shih-Fu ChangColumbia UniversityCT-ISG：Trustworthy Media：Blind Passive Media Forensics for Content Integrity Verification摘要：数字视听内容如今可以被修改和模拟，具有前所未有的真实感。为了应对恶意媒体操纵的挑战，我们建议开发新的理论，方法和一套全面的工具，可用于验证内容的完整性，在音频和视觉形式在不同的背景下，在多个层次。研究结果将在科学领域以及许多国家利益的实际应用中产生重大影响，例如值得信赖的新闻报道，监视安全，情报收集，刑事调查，金融交易和医疗信息管理。我们的努力将基于一个重要的范例，称为盲被动媒体取证，这从根本上不同于传统的方法，使用加密签名和水印。所提出的方法从信号和传感设备中提取潜在的、唯一的签名来检测篡改异常，使得验证仅需要手头的媒体内容而不需要任何附加信息或准备。具体来说，我们提出了一个系统的框架，设备签名一致性检查和联合多模态方法验证媒体的完整性。我们还将分析各种攻击场景，构建严格的基准数据集和性能指标，并采取积极措施，通过在线系统的公开测试来传播结果并提高对这一关键领域的认识。

Proposal: 0716203Shih-Fu ChangColumbia UniversityCT-ISG: Trustworthy Media: Blind Passive Media Forensics for Content Integrity VerificationAbstract:Digital audio-visual content can nowadays be modified and simulated with unprecedented realism. In response to the challenge of malicious media manipulation, we propose to develop new theories, methods, and a comprehensive suite of tools that can be used to verify content integrity in both audio and visual modalities at multiple levels under diverse contexts. The results will have significant impacts in scientific areas as well as many practical applications of national interest, such as trustworthy news reporting, surveillance security, intelligence gathering, criminal investigation, financial transactions, and medical information management. Our efforts will be based on an important paradigm, called blind passive media forensics, which fundamentally differs from conventional approaches using cryptographic signatures and watermarking. The proposed methods extract latent, unique signatures from the signals and sensing devices to detect tampering anomalies, so that verification requires only the media content at hand without any additional information or preparation. Specifically, we propose a systematic framework for device signature consistency checking and a joint multi-modal approach for verifying media integrity. We will also analyze various attack scenarios, construct rigorous benchmark datasets and performance metrics, and take proactive steps in disseminating results and promoting awareness of this critical area through public testing of online systems.