对于深度假冒检测,rPPG的本地关注和远距离互动
Jiahui Wu1, Yu Zhu1,2, Xiaoben Jiang1
1School of Information Science and Engineering, East China University of Science and Technology, Shanghai, 200237 China.
The Visual computer
|June 26, 2023
概括
这项研究引入了一种新的Deepfake检测方法,使用远程光电显微镜 (rPPG) 信号. 该方法通过分析面部血流中的独特节奏模式,有效地识别操纵的视频,优于现有的技术.
科学领域:
- 计算机视觉 计算机视觉
- 生物医学信号处理
- 人工智能的人工智能
背景情况:
- 深度假冒技术引起了公众的重大关注,因为它有可能被滥用.
- 现有的面部伪造检测方法需要改进,以应对复杂的操纵.
- 远程光电显微镜 (rPPG) 通过分析生理信号,为检测Deepfakes提供了一个有前途的途径.
研究的目的:
- 开发一种新的Deepfake检测方法,利用rPPG信号的独特节奏模式.
- 通过分析rPPG信号特征,将Deepfake检测视为源检测任务.
- 通过使用多尺度的时空信息,提高Deepfake检测的准确性和稳定性.
主要方法:
- 利用多尺度的时空PPG地图,从各种面部区域提取心跳信号.
- 提出了一个两阶段网络,包含一个面具引导局部注意 (MLA) 模块,用于本地PPG地图模式分析.
- 采用时间变压器来捕捉相邻的PPG地图中的远程空间和时间不一致.
主要成果:
- 与现有的基于rPPG的方法相比,该方法在FaceForensics++和Celeb-DF数据集上表现出卓越的性能.
- 实验证实了MLA模块和时间变压器在捕获歧视性特征方面的有效性.
- 视觉化验证了该方法能够检测出暗示Deepfakes的微妙文物的能力.
结论:
- rPPG信号为强大的Deepfake检测提供了强大的生物指标.
- 拟议的多尺度时空方法通过分析独特的生理信号模式,有效地识别Deepfakes.
- 这项工作推进了深度假冒检测的最新技术,提供了对操纵媒体的可靠防御机制.
更多相关视频
09:38Generalized Psychophysiological Interaction PPI Analysis of Memory Related Connectivity in Individuals at Genetic Risk for Alzheimer's Disease
Published on: November 14, 2017
15.0K
06:37Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
Published on: December 15, 2023
3.9K
相关概念视频
01:21Association Areas of the Cortex
5.6K
Association areas are regions of the cerebral cortex that do not have a specific sensory or motor function. Instead, they integrate and interpret information from various sources to enable higher cognitive processes such as memory, learning, and decision-making. Some key association areas include the following:
Prefrontal Association Area: This area is located in the frontal lobe and is involved in planning, decision-making, and moderating social behavior. It connects with primary motor areas,...
Prefrontal Association Area: This area is located in the frontal lobe and is involved in planning, decision-making, and moderating social behavior. It connects with primary motor areas,...
5.6K
01:20Parallel Processing
186
The brain processes sensory information rapidly due to parallel processing, which involves sending data across multiple neural pathways at the same time. This method allows the brain to manage various sensory qualities, such as shapes, colors, movements, and locations, all concurrently. For instance, when observing a forest landscape, the brain simultaneously processes the movement of leaves, the shapes of trees, the depth between them, and the various shades of green. This enables a quick and...
186