Exposing photo manipulation with inconsistent reflections

163 citations recorded by the source corpus.

How researchers see this paper

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

CNN-Generated Images Are Surprisingly Easy to Spot… for Now (2019, background): “Early work focused on hand-crafted cues [16] such as compression artifacts [5], resampling [37], or physical scene constraints [32].”
SafetyNet: Detecting and Rejecting Adversarial Examples Robustly (2017, background): “Current best methods to identify fake images require careful analysis of vanishing points [8], illumination angles [6], and shadows [11] (reviews in [8, 7]).”
Towards Universal Fake Image Detectors that Generalize Across Generative Models (2023, methodology): “When traditional methods are used to manipulate an image, the alteration in the underlying image statistics can be detected using hand-crafted cues such as compression artifacts [5], resampling [39] or irregular reﬂections [36].”
Detecting Photoshopped Faces by Scripting Photoshop (2019, background): “There has also been work that detects inconsistent quantization [2], double-JPEG artifacts [6, 3], and geometric inconsistencies [25].”
Exploiting Visual Artifacts to Expose Deepfakes and Face Manipulations (2019, background): “Examples for physics-based methods are inconsistencies in lighting [19] or reflections [17, 29].”