Exposing Deep Fakes Using Inconsistent Head Poses

Yang, Xin; Li, Yuezun; Lyu, Siwei

doi:10.1109/icassp.2019.8683164

Cited by 702 publications

(414 citation statements)

References 13 publications

Supporting

Mentioning

404

Contrasting

Unclassified

Order By: Relevance

“…Level 1 --four CNN models [6] Level 2 --MesoNet [9] Level 3 abnormal physiological signal breathing, pulse, eye blinking [11], head positions [13]…”

Section: Level Class Combination With Forensics Typical Examplesmentioning

confidence: 99%

See 1 more Smart Citation

A Novel Counterfeit Feature Extraction Technique for Exposing Face-Swap Images Based on Deep Learning and Error Level Analysis

Zhang

Zhao

2020

Entropy

View full text Add to dashboard Cite

The quality and efficiency of generating face-swap images have been markedly strengthened by deep learning. For instance, the face-swap manipulations by DeepFake are so real that it is tricky to distinguish authenticity through automatic or manual detection. To augment the efficiency of distinguishing face-swap images generated by DeepFake from real facial ones, a novel counterfeit feature extraction technique was developed based on deep learning and error level analysis (ELA). It is related to entropy and information theory such as cross-entropy loss function in the final softmax layer. The DeepFake algorithm is only able to generate limited resolutions. Therefore, this algorithm results in two different image compression ratios between the fake face area as the foreground and the original area as the background, which would leave distinctive counterfeit traces. Through the ELA method, we can detect whether there are different image compression ratios. Convolution neural network (CNN), one of the representative technologies of deep learning, can extract the counterfeit feature and detect whether images are fake. Experiments show that the training efficiency of the CNN model can be significantly improved by the ELA method. In addition, the proposed technique can accurately extract the counterfeit feature, and therefore achieves outperformance in simplicity and efficiency compared with direct detection methods. Specifically, without loss of accuracy, the amount of computation can be significantly reduced (where the required floating-point computing power is reduced by more than 90%).

show abstract

“…Level 1 --four CNN models [6] Level 2 --MesoNet [9] Level 3 abnormal physiological signal breathing, pulse, eye blinking [11], head positions [13]…”

Section: Level Class Combination With Forensics Typical Examplesmentioning

confidence: 99%

“…Li et al [12] used the color differences between GAN generated images and real images in a non-RGB color space to classify them. Xin et al [13] proposed a new method to detect and identify video forgery based on the inconsistency of the head positions. Peng et al [14] proposed a two-stream network for face tamper detection.…”

Section: Introductionmentioning

confidence: 99%

A Novel Counterfeit Feature Extraction Technique for Exposing Face-Swap Images Based on Deep Learning and Error Level Analysis

Zhang

Zhao

2020

Entropy

View full text Add to dashboard Cite

show abstract

“…Computer science approaches address a number of tasks that are important for fighting misinformation at scale, such as identifying deep fakes (Yang et al, 2018), analysing provenance and spread patterns (Wu et al, 2016), and assisting humans (e.g. fact-checkers and journalists) with ranking and prioritisation (Wiebe and Riloff, 2005;Cazalens et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

Understanding the Role of Human Values in the Spread of Misinformation

Farrell¹,

Piccolo²,

Perfumi³

et al. 2019

Proceedings of the Conference for Truth and Trust Online 2019

View full text Add to dashboard Cite

Social media platforms are often implicated in the spread of misinformation for encouraging the behaviour of rapid-sharing without adequate mechanisms for verifying information. To counter this phenomena, much related research in computer science has been focusing on developing tools to detect misinformation, to rank fact-check-worthy claims, and to understand their spread patterns, while psychosocial approaches have been focused on understanding information literacy, ideology and partisanship. In this paper, we demonstrate through a survey of nearly 100 people that the Human Values could have a significant influence on the way people perceive and share information. We argue that integrating a valuesoriented perspective into computational approaches for handling misinformation could encourage misinformation prevention, and assist in predicting and ranking misinformation.

show abstract

“…There are some related work proposed to detect AI generated fake images or videos using deep networks. To detect DeepFake video, different detection methods have been proposed [4,5,6,7,8]. In addition, some works focus on the detection of GAN generated images [9,10,11,12].…”

Section: Related Workmentioning

confidence: 99%

On the Generalization of GAN Image Forensics

Xuan

Peng

Wang

et al. 2019

Biometric Recognition

131

View full text Add to dashboard Cite

Recently GAN generated face images are more and more realistic with high-quality, even hard for human eyes to detect. On the other hand, the forensics community keeps on developing methods to detect these generated fake images and try to ensure the credibility of visual contents. Although researchers have developed some methods to detect generated images, few of them explore the important problem of generalization ability of forensics model. As new types of GANs are emerging fast, the generalization ability of forensics models to detect new types of GAN images is absolutely an essential research topic, which is also very challenging. In this paper, we explore this problem and propose to use preprocessed images to train a forensic CNN model. By applying similar image level preprocessing to both real and fake images, unstable low level noise cues are destroyed, and the forensics model is forced to learn more intrinsic features to classify the generated and real face images. Our experimental results also prove the effectiveness of the proposed method.

show abstract

Exposing Deep Fakes Using Inconsistent Head Poses

Cited by 702 publications

References 13 publications

A Novel Counterfeit Feature Extraction Technique for Exposing Face-Swap Images Based on Deep Learning and Error Level Analysis

A Novel Counterfeit Feature Extraction Technique for Exposing Face-Swap Images Based on Deep Learning and Error Level Analysis

Understanding the Role of Human Values in the Spread of Misinformation

On the Generalization of GAN Image Forensics

Contact Info

Product

Resources

About