Umur Aybars Ciftci

Title: Innovations of Umur Aybars Ciftci

Introduction

Umur Aybars Ciftci is an accomplished inventor based in Vestal, NY (US). He has made significant contributions to the field of synthetic content detection, particularly in identifying deep fakes through innovative methodologies. With a total of 2 patents, his work stands out for its practical applications and high accuracy.

Latest Patents

Ciftci's latest patent, titled "Fakecatcher: detection of synthetic portrait videos using biological signals," addresses the pressing issue of synthetic content in portrait videos. Traditional detectors that rely solely on deep learning techniques often fail to identify fake content, as generative models can produce highly realistic results. Ciftci's approach utilizes biological signals embedded in portrait videos, which are not preserved in fake content, serving as implicit descriptors of authenticity. His method achieves an impressive 99.39% accuracy in pairwise separation. By analyzing signal transformations and corresponding feature sets, he formulates a generalized classifier for fake content. The use of signal maps and a convolutional neural network (CNN) enhances the classifier's ability to detect synthetic content. Evaluations across various datasets demonstrate superior detection rates compared to baseline methods, regardless of the source generator or properties of the fake content. His experiments include signals from different facial regions, under various image distortions, and with different segment durations, showcasing the robustness of his approach.

Career Highlights

Ciftci has worked at the State University of New York, where he has contributed to research and development in his field. His innovative work has garnered attention and recognition, establishing him as a key figure in the realm of synthetic content detection.

Collaborations

Some of his notable coworkers include Ilke Demir and Lijun Yin, who have collaborated with him on various projects, enhancing the impact of his research.

Conclusion

Umur Aybars Ciftci's contributions to the detection of synthetic content through biological signals represent a significant advancement in the field. His innovative methodologies and high accuracy in detection highlight the importance of his work in combating the challenges posed by deep fakes.