2017
DOI: 10.1007/978-3-319-64185-0_9
|View full text |Cite
|
Sign up to set email alerts
|

Detection of Face Morphing Attacks by Deep Learning

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

2
71
0

Year Published

2018
2018
2021
2021

Publication Types

Select...
5
3
1

Relationship

1
8

Authors

Journals

citations
Cited by 93 publications
(73 citation statements)
references
References 20 publications
2
71
0
Order By: Relevance
“…Image morphing is a longstanding research topic in computer graphics and computer vision, e.g. [36,29,30,32,38,17]). Given two embedded images with their respective latent vectors w 1 and w 2 , morphing is computed by a linear interpolation, w = λw 1 + (1 − λ)w 2 , λ ∈ (0, 1), and subsequent image generation using the new code w. As Figure 4 shows, our method generates high-quality morphing between face images (row 1,2,3) but fails on non-face images in both in-class (row 4) and inter-class (row 5) morphing.…”
Section: Morphingmentioning
confidence: 99%
“…Image morphing is a longstanding research topic in computer graphics and computer vision, e.g. [36,29,30,32,38,17]). Given two embedded images with their respective latent vectors w 1 and w 2 , morphing is computed by a linear interpolation, w = λw 1 + (1 − λ)w 2 , λ ∈ (0, 1), and subsequent image generation using the new code w. As Figure 4 shows, our method generates high-quality morphing between face images (row 1,2,3) but fails on non-face images in both in-class (row 4) and inter-class (row 5) morphing.…”
Section: Morphingmentioning
confidence: 99%
“…In case of the deep learning techniques, we have used the pre-trained network and computed the corresponding features that are further classified using a linear Support Vector Machines (SVM). To this extent, we have considered pre-trained CNN such as AlexNet [11,33,34], GoogleNet [33], Inception V3 [33], ResNet101 [11,33,34], VGG16 [11,33,9] and VGG19 [11,33,9]. The deep-learning techniques are used only as the feature extraction techniques owing to the availability of the small datasets.…”
Section: Experiments and Resultsmentioning
confidence: 99%
“…Following this work, several other works are reported [21,7] using the capability of micro-texture extraction techniques that can effectively capture the variations to reflect the process of morphing, which aids the morph detection task. Lately, the use of pre-trained deep CNNs with different architectures are widely studied in [5,9,11]. Further, the combination of deep features with handcrafted features is proposed in [10].…”
Section: Related Workmentioning
confidence: 99%
“…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%