The Deepfake Detection Challenge (DFDC) Preview Dataset
read the original abstract
In this paper, we introduce a preview of the Deepfakes Detection Challenge (DFDC) dataset consisting of 5K videos featuring two facial modification algorithms. A data collection campaign has been carried out where participating actors have entered into an agreement to the use and manipulation of their likenesses in our creation of the dataset. Diversity in several axes (gender, skin-tone, age, etc.) has been considered and actors recorded videos with arbitrary backgrounds thus bringing visual variability. Finally, a set of specific metrics to evaluate the performance have been defined and two existing models for detecting deepfakes have been tested to provide a reference performance baseline. The DFDC dataset preview can be downloaded at: deepfakedetectionchallenge.ai
This paper has not been read by Pith yet.
Forward citations
Cited by 21 Pith papers
-
The Regularizing Power of Language-Training Deepfake Detectors
A dual-encoder deepfake detector pairs a frozen specialist with a LoRA-tuned MLLM, trained first via binary alignment then via RL to reward explain-then-classify behavior, yielding improved cross-dataset performance a...
-
Beyond Flicker: Detecting Kinematic Inconsistencies for Generalizable Deepfake Video Detection
The method decomposes facial landmark motions into bases with an autoencoder, selectively breaks their natural correlations to create kinematic inconsistencies, and uses these to train a detector that achieves state-o...
-
Chameleon: Benchmarking Detection and Backtracking on Commercial-Grade AI-Generated Videos
Chameleon is a new benchmark of commercial-grade AI videos for detection and forensic backtracking, showing existing methods struggle with high-fidelity spatiotemporally consistent content.
-
The DeepFake Detection Challenge (DFDC) Dataset
The DFDC dataset is the largest public collection of face-swapped videos and supports detectors that generalize to in-the-wild deepfakes.
-
When AUC Misleads: Polarization-Aware Evaluation of Deepfake Detectors under Domain Shift
Cross-AUC averages per-domain AUCs with a polarization term from Wasserstein distance on score distributions to assess deepfake detector generalization under domain shift more realistically than isolated AUC.
-
Venus-DeFakerOne: Unified Fake Image Detection & Localization
DeFakerOne integrates InternVL2 and SAM2 into a single model that achieves state-of-the-art results on 39 detection and 9 localization benchmarks for unified fake image detection and localization.
-
Rethinking Cross-Domain Evaluation for Face Forgery Detection with Semantic Fine-grained Alignment and Mixture-of-Experts
Cross-AUC exposes large robustness drops in existing face forgery detectors across datasets, while the SFAM model with semantic alignment and region-specific experts delivers better performance on public benchmarks.
-
Photonic Quantum Computing on Spin Memory Architecture with Tree-Encoded Fusion
Tree-encoded fusion with spin qubit memory suppresses erasure errors in photonic graph-state generation, yielding exponential compiler gains over OneAdapt on benchmarks and hardware validation.
-
Generalizable Face Forgery Detection via Separable Prompt Learning
A separable prompt learning strategy on CLIP's text encoder enables competitive or superior generalizable performance in cross-dataset and cross-method face forgery detection.
-
DeFakeQ: Enabling Real-Time Deepfake Detection on Edge Devices via Adaptive Bidirectional Quantization
DeFakeQ introduces an adaptive bidirectional quantization method tailored for deepfake detectors that maintains detection accuracy while enabling real-time performance on resource-constrained edge devices.
-
LAA-X: Unified Localized Artifact Attention for Quality-Agnostic and Generalizable Face Forgery Detection
LAA-X uses multi-task learning with explicit localized artifact attention and blending synthesis to build a deepfake detector that generalizes to high-quality and unseen manipulations after training only on real and p...
-
Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection
Orthogonal subspace decomposition via SVD on vision foundation model features preserves high-rank pre-trained knowledge by freezing principal components and adapting residuals, reducing overfitting for better generali...
-
Lightweight Complementary-Cue Fusion for Robust Video Face Forgery Detection
Lightweight fusion of WDF with SPSL or LBP cues into Xception improves AUC by 3.8-4.4% on FaceForensics++ and DFDC-Preview with negligible parameter overhead compared to larger frequency-based detectors.
-
Venus-DeFakerOne: Unified Fake Image Detection & Localization
DeFakerOne is a unified foundation model for joint image-level fake image detection and pixel-level localization that reports SOTA results on 39 detection and 9 localization benchmarks.
-
Photonic Quantum Computing on Spin Memory Architecture with Tree-Encoded Fusion
A new cross-dataset AUC metric exposes generalization gaps in deepfake detectors, and a CLIP-based framework with region-aware experts and patch-level alignment achieves state-of-the-art cross-domain robustness.
-
Tutor-Student Reinforcement Learning: A Dynamic Curriculum for Robust Deepfake Detection
A reinforcement learning tutor re-weights deepfake training samples using historical performance signals to improve generalization on unseen manipulations.
-
Generalizable Deepfake Detection Based on Forgery-aware Layer Masking and Multi-artifact Subspace Decomposition
FMSD improves cross-dataset generalization in deepfake detection by using gradient-based layer masking to select forgery-sensitive weights and SVD to split them into preserved semantic and multiple learnable artifact ...
-
Robust Deepfake Detection: Mitigating Spatial Attention Drift via Calibrated Complementary Ensembles
A multi-stream ensemble using DINOv2 and CLIP backbones trained with extreme degradations achieves stable deepfake detection and fourth place in the NTIRE 2026 challenge.
-
M3D-Net: Multi-Modal 3D Facial Feature Reconstruction Network for Deepfake Detection
M3D-Net reconstructs 3D facial features from RGB images and fuses them with RGB features through attention-based modules to achieve claimed state-of-the-art deepfake detection.
-
Face-D(^2)CL: Multi-Domain Synergistic Representation with Dual Continual Learning for Facial DeepFake Detection
Face-D²CL fuses spatial and frequency features and uses dual continual learning to reduce forgetting while adapting to new DeepFakes, cutting average error rates by 60.7% and raising unseen-domain AUC by 7.9% over prior SOTA.
-
Robust Deepfake Detection, NTIRE 2026 Challenge: Report
The NTIRE 2026 challenge finds that large foundation models combined with ensembles and degradation-aware training produce the most robust deepfake detectors.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.