Image Forgery Localization via Fine-Grained Analysis of CFA Artifacts

By Pasquale Ferrara, Tiziano Bianchi, Alessia De Rosa, Alessandro Piva. IEEE Transactions on Information Forensics and Security, 2012

485 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Law

View Paper: https://www.semanticscholar.org/paper/2fb36e3e7e97c51e0a50bf15b5c1e1697b9c08c2

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 Law

Citations in the corpus

Splicebuster: A new blind image splicing detector (2015, unknown): “[5].”
Face X-Ray for More General Face Forgery Detection (2019, background): “Early works [37, 8, 15] reveal the tampered regions using manually designed low-level image statistics at a local level.”
ForensicTransfer: Weakly-supervised Domain Adaptation for Forgery Detection (2018, background): “Traditional Media Forensics Traditional model-based approaches exploit a specific clue, like inconsistencies at pixel-level related for example to JPEG compression artifacts [4], demosaicking clues [16], lens aberration [41], or camera noise [29].”
Media Forensics and DeepFakes: An Overview (2020, background): “The problem can be also recast in a Bayesian framework, as proposed in [66], obtaining a probability map in output which allows for fine-grained localization of image tampering.”
Hybrid LSTM and Encoder–Decoder Architecture for Detection of Image Forgeries (2019, background): “, whether an image is tampered or not [13], [30], [39], [47], [65], [74], [78].”
Thinking in Frequency: Face Forgery Detection by Mining Frequency-aware Clues (2020, background): “, local pattern analysis [21], noise variances evaluation [47] and steganalysis features [13, 24] to discover forgery patterns and magnify faint discrepancy between real and forged images.”
Fighting Fake News: Image Splice Detection via Learned Self-Consistency (2018, methodology): “We compared our model with three methods that use image processing techniques to detect specific imaging artifacts: Color Filter Array (CFA) [44] detects artifacts in color pattern interpolation; JPEG DCT [45] detects inconsistencies over JPEG coefficients; and Noise Variance (NOI) [46] detects anom”
Two-Stream Neural Networks for Tampered Face Detection (2017, methodology): “[14] proposed an algorithm that estimates the camera ﬁl-ter pattern based on the fact that the variance of prediction error between CFA present regions (authentic regions) and CFA absent regions (tampered regions) is different.”
Learning Rich Features for Image Manipulation Detection (2018, methodology): “• CFA1:A CFA pattern estimation method [14] which uses nearby pixels to approximate the camera filter array patterns and then produces the tampering probability for each pixel.”
ManTra-Net: Manipulation Tracing Network for Detection and Localization of Image Forgeries With Anomalous Features (2019, methodology): “…ManTra-Net’s performance against numbers reported in [55], which provides scores of the classic unsupervised methods: ELA [29], NOI1 [34], CFA1 [22], and the latest DNN-based solutions, MFCN [39] and J-LSTM [7] on the four benchmark datasets, namely NIST 2016 [3], CASIA [2], COV-ERAGE [43],…”
Image Splicing Localization using a Multi-task Fully Convolutional Network (MFCN) (2017, background): “Following the acronyms in [37], these algorithms are: ADQ1 [23], ADQ2 [4], ADQ3 [1], NADQ [6], BLK [22], CFA1 [15], CFA2 [12], DCT [36], ELA [19], NOI1 [27], NOI2 [26] and NOI3 [9].”
PSCC-Net: Progressive Spatio-Channel Correlation Network for Image Manipulation Detection and Localization (2021, methodology): “Our baseline IMDL methods include ELA [30], NOI1 [42], CFA1 [15], J-LSTM [3], H-LSTM [4], RGBN [69], ManTra-Net [65], and SPAN [22] where SPAN has reported the SOTA performance on localization.”
Noiseprint: A CNN-Based Camera Model Fingerprint (2018, methodology): “Popular targets of such analyses are lens aberration [5]–[7], camera response function [8]–[10], color filter array (CFA) [11]–[13] or JPEG artifacts [14]–[16].”