A Deep Learning Approach to Universal Image Manipulation Detection Using a New Convolutional Layer

By Belhassen Bayar, Matthew Christopher Stamm. Information Hiding and Multimedia Security Workshop, 2016

817 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/b699b0033764214ad9486da966e1a1ad37932f6f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

What makes fake images detectable? Understanding properties that generalize (2020, background): “Other works have focused on training classifiers for the detection task, using a deep network either directly on RGB images [4,2,32] or alternative image representations [7,26].”
FaceForensics++: Learning to Detect Manipulated Facial Images (2019, methodology): “We are considering learning-based methods used in the forensic community for generic manipulation detection [9, 16], computer-generated vs natural image detection [49] and face tampering detection [4].”
Two-branch Recurrent Network for Isolating Deepfakes in Videos (2020, neutral): “69 [8] Bayar and Stamm 82.”
DeeperForensics-1.0: A Large-Scale Dataset for Real-World Face Forgery Detection (2020, background): “The benchmark includes six image-level face forgery detection baselines [6, 8, 12, 13, 15, 32].”
Multi-task Learning for Detecting and Segmenting Manipulated Facial Images and Videos (2019, methodology): “Other approaches to detection include using a constrained convolutional layer [8], using a statistical pooling layer [25], using a two-stream network [31], using a lightweight CNN network [2], and using two cascaded convolutional layers at the bottom of a CNN [23].”
Capsule-forensics: Using Capsule Networks to Detect Forged Images and Videos (2018, background): “Raghavendra et al. [23] described the special case of ﬁne-tuning two available CNNs while Rossler et al. [11] used only one CNN. Bayar and Stamm [24], Rah-mouni et al. [25], Afchar et al. [13], Quan et al. [26], and Li et al. [9] proposed their own networks.”
ManTra-Net: Manipulation Tracing Network for Detection and Localization of Image Forgeries With Anomalous Features (2019, background): “Most of the existing methods [9, 25, 37, 38, 49] only focus on image-level detection— whether or not an image is forged.”
Lips Don't Lie: A Generalisable and Robust Approach to Face Forgery Detection (2020, background): “Some earlier face forgery detection works bias the network away from learning highlevel features by constraining CNN filters [6] or using relatively shallow networks [3] to focus on mesoscopic features.”
DeepFake Detection by Analyzing Convolutional Traces (2020, background): “They addressed the problem of fake detection as a binary classification problem for each frame of manipulated videos, considering different techniques present in the state of the art [1, 4, 7, 8, 10, 34].”
A deep learning approach to detection of splicing and copy-move forgeries in images (2016, neutral): “[16] proposed a new convolutional layer in their CNN model to learn prediction”
Constrained Convolutional Neural Networks: A New Approach Towards General Purpose Image Manipulation Detection (2018, methodology): “well as a learning rate that decreases during training ( [26] uses a fixed learning rate) and train using a different batch size.”
Hybrid LSTM and Encoder–Decoder Architecture for Detection of Image Forgeries (2019, methodology): “Recent efforts, including [8], [9], [17], [62], [77] in the manipulation detection task, exploit deep learning based models.”
Media Forensics and DeepFakes: An Overview (2020, methodology): “In [136], instead, constrained filter weights are learnt during training, and the CNN is used also for classification.”
Deep Learning for Deepfakes Creation and Detection (2019, background): “This increases the pixel level statistical similarity between real images and fake images and requires the forensic classiﬁer to learn more intrinsic and meaningful features, which has better generalization capability than previous image forensics methods [51,52] or image steganalysis networks [53].”
Leveraging Frequency Analysis for Deep Fake Image Recognition (2020, methodology): “These traditional methods have recently been expanded by CNN-based methods (Bayar & Stamm, 2016; Bappy et al., 2017; Cozzolino et al., 2017; Zhou et al., 2018), which learn a more complex feature representation, improving the state-of-the-art for tampered media detection.”
Attributing Fake Images to GANs: Learning and Analyzing GAN Fingerprints (2018, background): “While recent state-of-the-art visual forensics techniques demonstrate impressive results for detecting fake visual media [16, 53, 27, 13, 22, 11, 35, 67, 68, 26], they have only focused on semantic, physical, or statistical inconsistency of specific forgery scenarios, e.”
FaceForensics: A Large-scale Video Dataset for Forgery Detection in Human Faces (2018, methodology): “…are no speciﬁc approaches in the current literature to detect Face2Face manipulations, we decided to consider learning-based methods used in the forensic community for generic manipulation detection [26,27], computer-generated vs natural image detection [39] and face tampering detection [41,42].”