Automatic Generation and Detection of Visually Faultless Facial Morphs

Makrushin, Andrey; Neubert, Tom; Dittmann, Jana

doi:10.5220/0006131100390050

Cited by 97 publications

(126 citation statements)

References 22 publications

Supporting

Mentioning

125

Contrasting

Order By: Relevance

“…Several researchers become aware of the danger of morphing attacks and developed different forensic methods to detect this kind of fraud. In contrast to the already proposed methods, which are based on image degeneration [4], Binarized Statistical Image Features (BSIF) [5], neural networks [6], [7] or JPEG compression artifacts [8], we propose a method that is based on a physical illumination model. Illumination estimation to detect frauds was already studied in detail by [9], [10] to detect compositions of multiple photographs.…”

Section: Introductionmentioning

confidence: 99%

Reflection Analysis for Face Morphing Attack Detection

Seibold

Hilsmann

Eisert

2018

2018 26th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

A facial morph is a synthetically created image of a face that looks similar to two different individuals and can even trick biometric facial recognition systems into recognizing both individuals. This attack is known as face morphing attack. The process of creating such a facial morph is well documented and a lot of tutorials and software to create them are freely available. Therefore, it is mandatory to be able to detect this kind of fraud to ensure the integrity of the face as reliable biometric feature. In this work, we study the effects of face morphing on the physically correctness of the illumination. We estimate the direction to the light sources based on specular highlights in the eyes and use them to generate a synthetic map for highlights on the skin. This map is compared with the highlights in the image that is suspected to be a fraud. Morphing faces with different geometries, a bad alignment of the source images or using images with different illuminations, can lead to inconsistencies in reflections that indicate the existence of a morphing attack.

show abstract

Section: Introductionmentioning

confidence: 99%

Reflection Analysis for Face Morphing Attack Detection

Seibold

Hilsmann

Eisert

2018

2018 26th European Signal Processing Conference (EUSIPCO)

View full text Add to dashboard Cite

show abstract

“…The morphing described in [1,4] are time consuming because it requires a manual retouch for more realistic appearance. To overcome this issue, Makrushin et al in [5] proposed an automatic splicing-based morphing algorithm to generate thousands of visually faultless facial morphs. In general, the quality of morphed images is 2-fold; (i) morphed images need to be visually faultless to human eyes (i.e.…”

Section: Introductionmentioning

confidence: 99%

“…These properties make splicing technique to pass the first morph quality measure, provided that both source images have similar skin color, but miserable regarding the second quality measure because splicing adopts geometry from one source image only and it may not look very similar to the other image. A combined morph technique was also proposed by [5] to overcome the limitations in the two previous algorithms. It warps the images into an average position first, then it cuts the facial regions, blending and finally stitching them back on to the warped image.…”

Section: Introductionmentioning

confidence: 99%

“…Different morphing techniques are expected to produce different changes on image features, and potential digital forensic schemes to detect morphing must identify the appropriate sensitive features and the nature of resulting changes in order to select appropriate classifiers. Makrushin et al in [5] proposed an automatic morph detector based on Benford features computed from quantized Discrete Cosine Transformation (DCT) coefficients of JPEG-Compressed images. Frank Benford's law, roughly, states that the frequency distribution of leading digits of a set of (natural) numbers is logarithmic.…”

Section: Introductionmentioning

confidence: 99%

“…Frank Benford's law, roughly, states that the frequency distribution of leading digits of a set of (natural) numbers is logarithmic. The morphing detector adopted in [5] is based on the hypothesis that unlike naturally generated data, manipulated data do not obey the Benford's law. Although high accuracy in detection of morphed random face images obtained using Benford's law, however, they note that using legitimate image processing techniques, one can create face images with similar Benford feature distribution [5,6].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Topological Data Analysis for Image Tampering Detection

Asaad

Jassim

2017

Digital Forensics and Watermarking

View full text Add to dashboard Cite

This paper introduces a topological approach to detection of image tampering for forensics purposes. This is based on the emerging Topological Data Analysis (TDA) concept of persistent homological invariants associated with certain image features. Image features of interest are pixels that have a uniform Local Binary pattern (LBP) code representing texture feature descriptors. We construct the sequence of simplicial complexes for increasing sequence of distance thresholds whose vertices are the selected set of pixels, and calculate the corresponding non-increasing sequence of homology invariants (number of connected components). The persistent homology of this construction describes the speed with which the sequence terminates, and our tamper detection scheme exploit its sensitivity to image tampering/degradation. We test the performance of this approach on a sufficiently large image dataset from a benchmark dataset of passport photos, and show that the persistent homology sequence defines a discriminating criterion for the morphing attacks (i.e. distinguishing morphed images from genuine ones).

show abstract

Improving face morph detection with the pairs training effect

Flack

Ritchie

Cartledge

et al. 2023

Applied Cognitive Psychology

View full text Add to dashboard Cite

It is becoming increasingly common for face morphs (weighted combinations of two people's photographs) to be submitted for inclusion in an official document, such as a passport. These images may sufficiently resemble both individuals that they can be used by either person in a ‘fraudulently obtained genuine’ document. Problematically, people are poor at detecting face morphs and there is limited evidence that this can be improved. Here, we tested whether the ‘pairs training effect’ (working in pairs, which we know improves unfamiliar face matching) can improve face morph detection. We found morph detection was more accurate when working in a pair. Further, the lower performer in the pair maintained this benefit when completing the task again individually. We conclude that the pairs training effect translates to face morph detection, and these findings have important implications for improving the detection of face morphs at the initial application stage.

show abstract

Automatic Generation and Detection of Visually Faultless Facial Morphs

Cited by 97 publications

References 22 publications

Reflection Analysis for Face Morphing Attack Detection

Reflection Analysis for Face Morphing Attack Detection

Topological Data Analysis for Image Tampering Detection

Improving face morph detection with the pairs training effect

Contact Info

Product

Resources

About