FaceForensics: A Large-scale Video Dataset for Forgery Detection in Human Faces

Abstract

With recent advances in computer vision and graphics, it is now possible to generate videos with extremely realistic synthetic faces, even in real time. Countless applications are possible, some of which raise a legitimate alarm, calling for reliable detectors of fake videos. In fact, distinguishing between original and manipulated video can be a challenge for humans and computers alike, especially when the videos are compressed or have low resolution, as it often happens on social networks. Research on the detection of face manipulations has been seriously hampered by the lack of adequate datasets. To this end, we introduce a novel face manipulation dataset of about half a million edited images (from over 1000 videos). The manipulations have been generated with a state-of-the-art face editing approach. It exceeds all existing video manipulation datasets by at least an order of magnitude. Using our new dataset, we introduce benchmarks for classical image forensic tasks, including classification and segmentation, considering videos compressed at various quality levels. In addition, we introduce a benchmark evaluation for creating indistinguishable forgeries with known ground truth; for instance with generative refinement models.

Video: https://youtu.be/Tle7YaPkO_k

Cited by in corpus (65)

• MesoNet: a Compact Facial Video Forgery Detection Network
• Media Forensics and DeepFakes: an overview
• FakeCatcher: Detection of Synthetic Portrait Videos using Biological Signals
• DeepFakes: a New Threat to Face Recognition? Assessment and Detection
• Deep Learning for Deepfakes Creation and Detection: A Survey
• Recurrent Convolutional Strategies for Face Manipulation Detection in Videos
• Deepfake Detection by Human Crowds, Machines, and Machine-informed Crowds
• ForensicTransfer: Weakly-supervised Domain Adaptation for Forgery Detection
• The Deepfake Detection Challenge (DFDC) Preview Dataset
• Use of a Capsule Network to Detect Fake Images and Videos
• Video-to-Video Synthesis
• Train Sparsely, Generate Densely: Memory-efficient Unsupervised Training of High-resolution Temporal GAN
• Celeb-DF: A Large-scale Challenging Dataset for DeepFake Forensics
• Emotional Speech-Driven Animation with Content-Emotion Disentanglement
• Deep Spatial Transformation for Pose-Guided Person Image Generation and Animation
• More Real than Real: A Study on Human Visual Perception of Synthetic Faces
• Deepfake: Definitions, Performance Metrics and Standards, Datasets and Benchmarks, and a Meta-Review
• Deferred Neural Rendering: Image Synthesis using Neural Textures
• Learnable Gated Temporal Shift Module for Deep Video Inpainting
• On the Detection of Digital Face Manipulation
• Multi-task Learning For Detecting and Segmenting Manipulated Facial Images and Videos
• Attributing Fake Images to GANs: Learning and Analyzing GAN Fingerprints
• We Need No Pixels: Video Manipulation Detection Using Stream Descriptors
• Deep Convolutional Neural Network for Identifying Seam-Carving Forgery
• DeepFake Detection Based on the Discrepancy Between the Face and its Context
• Detecting Photoshopped Faces by Scripting Photoshop
• Deep Image Spatial Transformation for Person Image Generation
• Hiding Faces in Plain Sight: Disrupting AI Face Synthesis with Adversarial Perturbations
• Face X-ray for More General Face Forgery Detection
• Imperfect ImaGANation: Implications of GANs Exacerbating Biases on Facial Data Augmentation and Snapchat Selfie Lenses
• A Convolutional LSTM based Residual Network for Deepfake Video Detection
• Frame-Recurrent Video Inpainting by Robust Optical Flow Inference
• Exposing GAN-synthesized Faces Using Landmark Locations
• Vulnerability of Face Recognition to Deep Morphing
• Example-Guided Style Consistent Image Synthesis from Semantic Labeling
• Live Speech Portraits: Real-Time Photorealistic Talking-Head Animation
• Deepfake detection: humans vs. machines
• Geo-DefakeHop: High-Performance Geographic Fake Image Detection
• Exploiting Human Social Cognition for the Detection of Fake and Fraudulent Faces via Memory Networks
• Video Inpainting by Jointly Learning Temporal Structure and Spatial Details
• Free-form Video Inpainting with 3D Gated Convolution and Temporal PatchGAN
• How Do the Hearts of Deep Fakes Beat? Deep Fake Source Detection via Interpreting Residuals with Biological Signals
• MotionVideoGAN: A Novel Video Generator Based on the Motion Space Learned from Image Pairs
• On the generalization of GAN image forensics
• Deepfake Video Forensics based on Transfer Learning
• FDFtNet: Facing Off Fake Images using Fake Detection Fine-tuning Network
• FLNet: Landmark Driven Fetching and Learning Network for Faithful Talking Facial Animation Synthesis
• Few-shot Video-to-Video Synthesis
• Fake face detection via adaptive manipulation traces extraction network
• Headset: Human emotion awareness under partial occlusions multimodal dataset
• A Novel BiLevel Paradigm for Image-to-Image Translation
• Dense Pose Transfer
• FReTAL: Generalizing Deepfake Detection using Knowledge Distillation and Representation Learning
• Example-Guided Image Synthesis across Arbitrary Scenes using Masked Spatial-Channel Attention and Self-Supervision
• Live Face De-Identification in Video
• Designing Indicators to Combat Fake Media
• Expression-aware video inpainting for HMD removal in XR applications
• Example-Guided Scene Image Synthesis using Masked Spatial-Channel Attention and Patch-Based Self-Supervision
• CASIA-Face-Africa: A Large-scale African Face Image Database
• Detecting Forged Facial Videos using convolutional neural network
• How Relevant is the Turing Test in the Age of Sophisbots?
• Impact of Benign Modifications on Discriminative Performance of Deepfake Detectors
• Deepfake Representation with Multilinear Regression
• One Shot Face Swapping on Megapixels
• Learning Transformation-Aware Embeddings for Image Forensics