{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/deepfakes-a-new-threat-to-face-recognition","title":"DeepFakes: a New Threat to Face Recognition? Assessment and Detection","arxiv_id":"1812.08685","date":"2018-12-20","proceeding":null,"authors":["Pavel Korshunov","Sebastien Marcel"],"abstract":"It is becoming increasingly easy to automatically replace a face of one\nperson in a video with the face of another person by using a pre-trained\ngenerative adversarial network (GAN). Recent public scandals, e.g., the faces\nof celebrities being swapped onto pornographic videos, call for automated ways\nto detect these Deepfake videos. To help developing such methods, in this\npaper, we present the first publicly available set of Deepfake videos generated\nfrom videos of VidTIMIT database. We used open source software based on GANs to\ncreate the Deepfakes, and we emphasize that training and blending parameters\ncan significantly impact the quality of the resulted videos. To demonstrate\nthis impact, we generated videos with low and high visual quality (320 videos\neach) using differently tuned parameter sets. We showed that the state of the\nart face recognition systems based on VGG and Facenet neural networks are\nvulnerable to Deepfake videos, with 85.62% and 95.00% false acceptance rates\nrespectively, which means methods for detecting Deepfake videos are necessary.\nBy considering several baseline approaches, we found that audio-visual approach\nbased on lip-sync inconsistency detection was not able to distinguish Deepfake\nvideos. The best performing method, which is based on visual quality metrics\nand is often used in presentation attack detection domain, resulted in 8.97%\nequal error rate on high quality Deepfakes. Our experiments demonstrate that\nGAN-generated Deepfake videos are challenging for both face recognition systems\nand existing detection methods, and the further development of face swapping\ntechnology will make it even more so.","url_abs":"http://arxiv.org/abs/1812.08685v1","url_pdf":"http://arxiv.org/pdf/1812.08685v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"deepfakes-a-new-threat-to-face-recognition","repo_url":"https://github.com/CatoGit/Comparing-the-Performance-of-Deepfake-Detection-Methods-on-Benchmark-Datasets","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"paper_slug":"deepfakes-a-new-threat-to-face-recognition","repo_url":"https://github.com/Recognito-Vision/Linux-FaceRecognition-FaceLivenessDetection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"constrained-lip-synchronization","task_name":"Constrained Lip-synchronization"},{"task_slug":"face-recognition","task_name":"Face Recognition"},{"task_slug":"face-swapping","task_name":"Face Swapping"},{"task_slug":null,"task_name":"Generative Adversarial Network"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1812.08685","atlas_url":"https://app.syntology.ai/?focus=1812.08685","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}