DeepfakeTIMIT

DeepfakeTIMIT is a database of videos where faces are swapped using the open source GAN-based approach (adapted from here: https://github.com/shaoanlu/faceswap-GAN), which, in turn, was developed from the original autoencoder-based Deepfake algorithm.

When creating the database, we manually selected 16 similar looking pairs of people from publicly available VidTIMIT database. For each of 32 subjects, we trained two different models: a lower quality (LQ) with 64 x 64 input/output size model, and higher quality (HQ) with 128 x 128 size model (see the available images for the illustration). Since there are 10 videos per person in VidTIMIT database, we generated 320 videos corresponding to each version, resulting in 620 total videos with faces swapped. For the audio, we kept the original audio track of each video, i.e., no manipulation was done to the audio channel.

Any publication (eg. conference paper, journal article, technical report, book chapter, etc) resulting from the usage of DeepfakeTIMIT must cite the following paper:

    P. Korshunov and S. Marcel,
    DeepFakes: a New Threat to Face Recognition? Assessment and Detection.
    arXiv and Idiap Research Report

Any publication (eg. conference paper, journal article, technical report, book chapter, etc) resulting from the usage of VidTIMIT and subsequently DeepfakeTIMIT must also cite the following paper:

    C. Sanderson and B.C. Lovell,
    Multi-Region Probabilistic Histograms for Robust and Scalable Identity Inference.
    Lecture Notes in Computer Science (LNCS), Vol. 5558, pp. 199-208, 2009.

DeepfakeTIMIT是一类基于开源生成对抗网络（GAN）实现人脸交换的视频数据库（改编自https://github.com/shaoanlu/faceswap-GAN），而该开源方案本身源自最初基于自编码器的Deepfake算法。

在构建该数据库过程中，研究人员从公开可用的VidTIMIT数据库中手动遴选了16对外貌相似的人物。针对VidTIMIT数据库中的32位受试者，我们分别训练了两种模型：输入输出分辨率为64×64的低质量（LQ）模型，以及分辨率为128×128的高质量（HQ）模型（详见配套示例图像）。由于VidTIMIT数据库中每位受试者对应10段视频，我们针对每个模型版本各生成了320段人脸交换视频，最终总计得到620段人脸交换视频。音频层面，我们保留了每段视频的原始音轨，未对音频通道进行任何修改。

凡使用DeepfakeTIMIT数据库产出的任何形式的出版物（包括但不限于会议论文、期刊文章、技术报告、图书章节等），均需引用以下论文：
P. Korshunov与S. Marcel，
《DeepFakes: a New Threat to Face Recognition? Assessment and Detection》，
arXiv及Idiap研究报告

凡同时使用VidTIMIT数据库及后续衍生的DeepfakeTIMIT数据库产出的任何形式的出版物，亦需引用以下论文：
C. Sanderson与B.C. Lovell，
《Multi-Region Probabilistic Histograms for Robust and Scalable Identity Inference》，
《计算机科学讲义》（Lecture Notes in Computer Science, LNCS），第5558卷，第199-208页，2009年。