MUMBAI, India, July 31 -- Intellectual Property India has published a patent application (202411102592 A) filed by Lovely Professional University on December 24, 2024, for Advanced Biometric Facial Recognition System For Identifying Individuals With Highly Similar Facial Features.
Inventors include Krishnendu Das; Dr. Krishan Arora; and Dr. Himanshu Sharma.
The application for the patent was published on July 24, 2026, under issue no. 30/2026.
Abstract: ABSTRACT ADVANCED BIOMETRIC FACIAL RECOGNITION SYSTEM FOR IDENTIFYING INDIVIDUALS WITH HIGHLY SIMILAR FACIAL The DetecTwin system is an advanced biometric facial recognition solution designed to overcome the challenges of identifying individuals with highly similar facial features, such as identical twins or doppelgangers. Traditional facial recognition systems often struggle to differentiate between individuals with near-identical faces, leading to security risks and identity verification issues. DetecTwin addresses this limitation by employing a Siamese neural network to compare facial image pairs, improving the system's ability to detect subtle facial differences. The system also incorporates attention mechanisms to focus on critical facial regions (e.g., eyes, nose, mouth), enhancing recognition accuracy. Additionally, DetecTwin integrates multi-modal biometric data, such as iris scans, to improve robustness and reliability. Data augmentation techniques further enrich the training dataset, allowing the system to perform well in real-world conditions. Designed with strong ethical considerations, DetecTwin ensures privacy and compliance with legal standards. This solution provides a reliable, high-accuracy approach for identity verification in security- sensitive applications.
Disclaimer: Curated by HT Syndication.
