How realistic is photorealistic?

By Siwei Lyu, Hany Farid. IEEE Transactions on Signal Processing, 2005

360 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/be5992ce7a8ac92f96993873590b673dca34c8d5

How researchers see this paper

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

Citations in the corpus

An improved approach to steganalysis of JPEG images (2010, methodology): “These images are first cropped into 256 256 pixels according to the method of [15,16,19,21], in order to remove the low complexity parts (all original uncompressed images are available at http://www.”
Media Forensics and DeepFakes: An Overview (2020, background): “Therefore, the most effective features are typically extracted from noise residuals, either in the spatial [114], [115] or in a transform domain [116], [117].”
Digital image forgery detection using passive techniques: A survey (2013, methodology): “Lyu and Farid (2005) created a statistical model for photographic images consisting of first and higher-order wavelet statistics using LDA and a non-linear SVM.”
The Design of High-Level Features for Photo Quality Assessment (2006, background): “For example, images are classified as photos or graphics [2, 7], taken indoors or outdoors [9, 14], city or landscape [18], photos or paintings [4], and real or rendered photo [10].”
Feature mining and pattern classification for steganalysis of LSB matching steganography in grayscale images (2008, methodology): “[13] S. Lyu, H. Farid, How realistic is photorealistic, IEEE Trans.”
Automatic fruit and vegetable classification from images (2010, background): “, 1997); low- and middle-level features to d istinguish broad classes of images (Rocha and Goldenstein, 20 07; Lyu and Farid, 2005; Cutzu et al., 2005; Serrano et al., 2004) ; In addition, Heidemann (2004) has presented an approach to e stablish image categories automatically using histograms, color”
Noiseprint: A CNN-Based Camera Model Fingerprint (2018, methodology): “the set of wavelet-based features of [36] is augmented with prediction error features computed on a noise residual extracted through denoising.”
Region Duplication Detection Using Image Feature Matching (2010, background): “Other intrinsic cues from the imaging process, including color filter array interpolation patterns [44], [46], camera response functions [30], and camera sensor noise [9], [18], or general statistical properties of untampered natural images [34],”
Detection of GAN-Generated Fake Images over Social Networks (2018, background): “Some papers propose to detect computer graphics images based on statistics extracted from their wavelet decomposition [5], or from residual images [6].”
Image forgery detection (2009, methodology): “In both cases, the statistical model is used to detect everything from basic image manipulations such as resizing and filtering [1] to discriminating photographic from computer-generated images [29] and detecting hidden messages (steganography) [30] .”
Using Color Compatibility for Assessing Image Realism (2007, background): “Farid and colleagues have done extensive work on using higher-order image statistics for a variety of tasks, including distinguishing between computer renderings and photographs [12], detecting digital tampering, finding art fakes [13], etc.”
Digital image forensics via intrinsic fingerprints (2008, background): “In these processing stages, each camera uses a different algorithm (that may be proprietary to the camera manufacturer, brand, or model) and leaves intrinsic fingerprint traces on the output data.”
Information Forensics: An Overview of the First Decade (2013, background): “Forensic techniques to identify forgeries using these fingerprints, known as photo-response non-uniformity (PRNU) fingerprints, will be discussed in detail in Section III.”
Detecting Digital Image Forgeries by Measuring Inconsistencies of Blocking Artifact (2007, background): “The second attempt is based on the statistics of natural image [5], inconsistencies based on scene lighting direction [6], or camera response normality [7].”