サイバーリンク Company Overview

■ FaceMe Security "Visitor Access Management" Main Features and Characteristics < Set Access Period for Each Visitor > - Individual settings for the dates and times visitors can enter - Access periods can be optimized according to the visitor's situation, whether for short visits or long stays. < Two Types of Authentication Methods to Accommodate Various Scenarios > ▶ Authentication via IP Camera - Enables authentication from a distance, achieving facial recognition without the need to disembark. ▶ Authentication via Facial Recognition Device (tablets, etc.) - Ideal for entry scenes such as reception or security posts. Can be flexibly utilized depending on the installation location and operation. - Allows for on-the-spot registration of facial information using the device, enabling smooth responses to sudden visits or unexpected guests. < Real-Time Notifications Supporting Security and Reception Operations > - Real-time notifications of visitor and employee classifications, authorization expiration, and authentication status. - Prevents delays and oversights in verification, balancing the reduction of security staff workload with enhanced security. These features resolve issues arising from visual checks, logbooks, and IC card operations, simultaneously achieving enhanced security and improved efficiency in reception and security operations.

**Improvement of Spoofing Prevention Accuracy and Performance** With FaceMe SDK 7.8, we enhance measures against increasingly sophisticated spoofing attacks by improving authentication accuracy and processing speed, thereby increasing the safety and reliability of facial recognition. - IR + RGB Spoofing Prevention: Detection accuracy improved by up to 35%, processing speed improved by up to 15% - 3D Spoofing Prevention: Detection accuracy improved by up to 56.8% for Intel RealSense D415 **Reducing Development Burden with the New Integrated API "FaceMe Authenticator"** We have integrated multiple processes that previously required individual implementation into a single API. Integrated Functions: - Face Detection + Feature Point Extraction + Spoofing Prevention + Authentication Processing Features: - Supports Windows / Ubuntu x64 / Android - Allows use of 2D cameras, 3D depth cameras, and IR + RGB cameras through a single interface - Significantly reduces development load while maintaining high-accuracy authentication

CyberLink has announced the start of a no-code integration between its AI facial recognition security system "FaceMe Security" and Asteria's no-code AI/IoT platform "Gravio." ■ Features Realized through the Integration of "FaceMe Security" and "Gravio" - Balancing Privacy Protection and High-Speed Processing Facial images are analyzed on an edge PC without being sent to the cloud, ensuring privacy protection through on-memory processing. - High-Precision, Low-Latency Walkthrough Authentication Possible While Walking Thanks to high-speed processing at the edge, users can smoothly undergo facial recognition without having to stop. - Flexible Integration with Various External Services through No-Code Gravio integrates with over 60 external services (such as Slack, Teams, kintone, BI tools, etc.) through no-code, enabling immediate notification of authentication results and data analysis, thus achieving automation tailored to the field. - Realization of Short-Term, Low-Cost Implementation No-code construction eliminates the need for dedicated development, allowing for the quick implementation of the facial recognition system without requiring large-scale infrastructure, thus enabling low-cost deployment.