YUAN Excels in NIST FRTE 1:N Rankings and Advances Face Recognition Technology


YUAN continues to dominate the field of facial recognition technology in Taiwan, according to a recent report. The FRTE ( Face Recognition Technology Evaluation ) 1:1 algorithm maintains the top spot in both the VISA/BORDER and BORDER/KIOSK test sets. In the VISA/BORDER set, YUAN achieved an exceptional 99.72% accuracy, showcasing its superior capabilities in one-to-one facial recognition.

Furthermore, YUAN's performance in the FRTE1:N algorithm is equally impressive, securing the top position in Taiwan across both VISA/BORDER and BORDER/KIOSK sets. Notably, our FRTE1:N algorithm achieved a remarkable 98.57% accuracy in the VISA/BORDER test set, once again demonstrating its outstanding performance in handling large-scale facial recognition scenarios.

This accomplishment is highlighted by the extensive VISA/BOARD datasets, featuring 1,000,000 pairs of images from 100 countries. Algorithms are meticulously ranked within the VISA/BORDER dataset, calculated using FNMR @ FMR = 10^(-6).

The VISA/BORDER combination serves as the default benchmark for leaderboards, replicating real-world scenarios involving visa submission and identity verification at border checkpoints. It offers a comprehensive evaluation that aligns with the intricacies of international travel and immigration procedures.

The BORDER/KIOSK, part of Unconstrained testing, is crucial for refining technology. Unconstrained face recognition faces challenges in real-world complexities, requiring robust algorithms for accurate identification. Despite difficulties, it's vital for enhancing security, driving innovation, and modernizing public services across sectors. Overcoming these challenges highlights its significance in shaping future security and technology.

Committed to excellence, YUAN High-Tech continually invests in optimizing algorithms. Our dedication to improvement reflects our desire for customers to share in the joy of our success.Our success is thanks to the hard work of the NexVDO SDK team, and we extend gratitude to our customers. Together, we anticipate achieving more milestones in video capture modules and NexVDO SDK services.  

Please visit : NexVDO SDK