Smart VKYC with Deepfake and Liveness Detection
DOI:
https://doi.org/10.63282/3117-5481/AIJCST-V8I3P108Keywords:
Artificial Intelligence, Video Know Your Customer, Deepfake Detection, Liveness Detection, OCR, Identity VerificationAbstract
The fast growth of virtual banking has multiplied the demand for at ease, remote, and seamless patron onboarding, making Video-primarily based KYC (vKYC) a essential procedure. This painting provides an incorporated AI-pushed verification device called clever vKYC, designed to automate identification validation and save you rising fraud tries in actual time. The goal of the system is to bolster virtual accept as true with by means of combining multiple gadget mastering and computer-imaginative and prescient modules into an stop-to-cease verification pipeline. The methodology includes FastAPI for high-performance backend processing, Agora for actual-time video streaming, Dlib’s 68-factor landmark model for blink-primarily based liveness detection, and a transformer-based totally deepfake detection version to perceive synthetic media. additionally, a twin-layer document validation method is implemented, in which Tesseract OCR extracts Aadhaar information and Pyzbar verifies cryptographically signed UIDAI QR codes to discover tampered or forged documents. Experimental consequences display a full-size improvement in preventing identity spoofing when FaceNet512 embeddings, OCR-QR consistency assessments, and deepfake ratings are mutually evaluated. The decision engine integrates output from these modules to generate a reliable confidence score, ensuring accurate verification without compromising person enjoy. This look at demonstrates that a multi-component vKYC architecture can considerably enhance the security and efficiency of far off consumer onboarding systems.
References
[1] Jain S, Gupta S. The future of digital banking: Video KYC and the rise of remote identity verification. IEEE Trans Comput Soc Syst. 2023;10(2):45–58. doi: 10.1109/TCSS.2023.3245678.
[2] Yu Z, Qin J, Li X, Zhao C, Lei Z, Zhao G. Deep learning for face anti-spoofing: A survey. IEEE Trans Pattern Anal Mach Intell. 2023;45(5):5609–5631. doi: 10.1109/TPAMI.2022.3149123.
[3] Soukupová T, Čech J. Real-time eye blink detection using facial landmarks. In: Proceedings of the 21st Computer Vision Winter Workshop (CVWW); 2016. p. 1–8.
[4] Smith R. An overview of the Tesseract OCR engine. In: Proceedings of the Ninth International Conference on Document Analysis and Recognition (ICDAR); 2007. Vol. 2. p. 629–633. doi: 10.1109/ICDAR.2007.4376991.
[5] Tolosana R, Vera-Rodriguez R, Fierrez J, Morales A, Ortega-Garcia J. Deepfakes and beyond: A survey of face manipulation and fake detection. Inf Fusion. 2020;64:131–148. doi: 10.1016/j.inffus.2020.06.014.
[6] Rossler A, Cozzolino D, Verdoliva L, Riess C, Thies J, Niessner M. FaceForensics++: Learning to detect manipulated facial images. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV); 2019. p. 1–11. doi: 10.1109/ICCV.2019.00009.
[7] Li Y, Yang X, Sun P, Qi H, Lyu S. Celeb-DF: A large-scale challenging dataset for deepfake forensics. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR); 2020. p. 3207–3216. doi: 10.1109/CVPR42600.2020.00327.
[8] Lyu S. Deepfake detection: Current challenges and next steps. In: IEEE International Conference on Multimedia and Expo Workshops (ICMEW); 2020. p. 1–6. doi: 10.1109/ICMEW46912.2020.9105991.
[9] Reserve Bank of India. Master direction – Know Your Customer (KYC) direction, 2016 (Updated 2023). RBI Notifications; 2023.
