| 摘要: |
| 目的 针对目前缺乏一种能够在复杂场景中暴露假脸图像的强大的假人脸检测模型,提出了一种新的网络,
称为双流操纵痕迹网络( Two-stream Manipulation Trace Network, TSMTN) ,用于学习假图像面部区域上的细微操纵
痕迹。 方法 该方法不同以往直接从图像中学习特征,而是先从图像中提取操纵痕迹,之后通过操纵痕迹检测人脸
是否被操纵。 该网络由于 3 个关键模块组成:空间域操纵痕迹提取( Spatial Domain Manipulation Trace Extraction,
SDMTE) 、频域操纵痕迹提取( Frequency Domain Manipulation Trace Extraction, FDMTE) 以及基于自注意力机制的
特征融合模块( Feature Fusion Module, FFM) 。 SDMTE 使用卷积神经网络( CNNs) 来学习图像空间域中的细微操纵
痕迹。 FDMTE 学习图像频域中高频信息的操纵痕迹。 FFM 融合空间域和频域中的操纵痕迹,以生成用于分类的
最终特征。 结果 实验结果表明:该模型具有良好的性能,在常用检测数据集上到达了先进的水平。 结论 该方法表
现出较好的鲁棒性和泛化能力,取得了一些进步,具有重要意义。 |
| 关键词: 人脸检测 操纵痕迹 空间域 频域 自注意力机制 |
| DOI： |
| 分类号: |
| 基金项目: |
|
| Face Forgery Detection Method Based on Manipulation Trace Fusion |
|
HUANG Jisheng YANG Gaoming
|
|
School of Computer Science and Engineering Anhui University of Science and Technology Anhui Huainan 232001
China
|
| Abstract: |
| Objective In view of the current lack of a powerful fake face detection model that can expose fake face images
in complex scenes a new network called two-stream manipulation trace network TSMTN is proposed for learning subtle
manipulation traces on facial regions in fake images. Methods This method is different from the previous methods of
directly learning features from images. Instead it first extracts manipulation traces from the image and then uses the
manipulation traces to detect whether the face has been manipulated. The network consists of three key modules spatial
domain manipulation trace extraction SDMTE frequency domain manipulation trace extraction FDMTE and feature
fusion module FFM based on self-attention mechanism. SDMTE uses convolutional neural networks CNNs to learn
subtle manipulation traces in the image spatial domain. FDMTE learns the manipulation traces of high-frequency
information in the frequency domain of images. FFM fuses manipulation traces in the spatial and frequency domains to
generate final features for classification. Results The experimental results show that the model has good performance and
has reached an advanced level on commonly used detection datasets. Conclusion This method shows good robustness and
generalization ability and has made some progress which is of great significance. |
| Key words: face detection manipulation traces spatial domain frequency domain self-attention mechanism |
