Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{192100,
        author = {Sonali and Mohd. Aman and Shadan Saifi},
        title = {An AI-Based Framework for Age-Invariant Face Recognition in Missing Person Identification},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {9},
        pages = {512-519},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=192100},
        abstract = {Identifying missing persons over long time intervals is a critical application of biometric systems, but facial aging poses a major challenge for recognition accuracy. In this paper, we propose a comprehensive AI-driven framework that integrates deep face embedding with age-progression modeling and fast similarity search to mitigate the effects of age variation. We employ a state-of-the-art face embedding network (ArcFace) to extract discriminative features, and augment the reference database with synthetic aged face images generated by a conditional adversarial autoencoder. Retrieval is performed with Facebook AI Similarity Search (Faiss) for efficient nearest-neighbor matching. Our experiments on benchmark aging datasets demonstrate that incorporating age-progressed faces significantly improves identification accuracy and F1-score (e.g., a ~3% accuracy gain over baseline) for large age gaps[1][2]. The proposed system effectively retrieves the correct identity despite age differences, illustrating its potential for aiding long-term missing person searches.},
        keywords = {Age-invariant face recognition, missing person identification, ArcFace, conditional adversarial autoencoder, FAISS, face embedding, age-progression.},
        month = {February},
        }