Discriminating between computer-generated images (CGIs) and photographic images (PIs) is not a new problem in digital image forensics. However, with advances in rendering techniques supported by strong hardware and in generative adversarial networks, CGIs are becoming indistinguishable from PIs in both human and computer perception. This means that malicious actors can use CGIs for spoofing facial authentication systems, impersonating other people, and creating fake news to be spread on social networks. The methods developed for discriminating between CGIs and PIs quickly become outdated and must be regularly enhanced to be able to reduce these attack surfaces. Leveraging recent advances in deep convolutional networks, we have built a modular CGI-PI discriminator with a customized VGG-19 network as the feature extractor, statistical convolutional neural networks as the feature transformers, and a discriminator. We also devised a probabilistic patch aggregation strategy to deal

Computer-based automatically generated text are used in various applications (e.g. text summarization, machine translation) and such the machine-generated text significantly helps our social life. However, machine-generated text may produce confusing information sometimes due to errors or inappropriate use of wordings caused by language processing, which could be a critical issue in president elections or in product advertisements. Previous methods for detecting such machinegenerated text typically estimates the text fluency, but, this may not be useful in near future because recently proposed neuralnetwork based natural language generation results in improved wording close to human-crafted one. However, we hypothesize that the habit of human on writing is still more consistent. For instance, the Zipf's law states that the most frequent word in the text written by human approximates twice the second most frequent word, nearly three times the third most frequent word, and so forth. We found that this is not true in the case of machine-generated text. We hence propose a method to identify the machine-generated text based on such the statistics-First, word distributed frequencies are compared with the Zipfian distribution to extract frequency features. Second, complex phrase features are extracted to show that humangenerated text contains more sophisticated phrases than machinegenerated one. Finally, the higher consistency of the humangenerated text is quantified at both the sentence level using phrasal verbs and at the paragraph level based on coreference resolution relationships, which are integrated into consistency features. The combination of the frequency, the complex phrase, and the consistency features is evaluated on a hundred of original English books and a hundred of translated ones from Finnish. The result shows that our method achieves the better performance (accuracy = 98.0% and equal error rate = 2.9%) comparing with a state-of-the-art method using parsing tree feature extraction. An advantage of this method is that this method can be used for large collections of text such as books efficiently. Other evaluation results in two other languages including French and Dutch showed similar results. They demonstrated that the proposed method works consistently in various languages.

The human gait has become another biometric trait used in security systems because it is unique to each person and can be recognized at a distance. However, a bad actor could use a gait recognition system to identify a person on the basis of his or her gait. We have developed a gait anonymization method that prevents unauthorized gait recognition. It modifies the gait so that the person cannot be identified while maintaining the naturalness of the gait. The modification is done by adding another gait, called "noise gait". A convolutional neural network makes this modification by taking two gaits as input, the original gait and the noise gait, and outputting an anonymized gait. The proposed method was evaluated using the success rate and mean opinion score (MOS). The success rate is the rate of failed gait recognition, and the MOS is a measure of the naturalness of the anonymized gait. In our experiments, the success rate achieved 98.86% at most while the highest naturalness score is 3.73 in the MOS scale. These findings should open new research directions regarding privacy protection related to gait recognition.

No abstract

No abstract