Wasseem N. Ibrahem Al-Obaydy and Shahrel Azmin Suandi
الملخص الانجليزي
Face recognition has received a substantial attention by the vision community over the past few decades. Most of the proposed frameworks have adopted the closed-set form of face recognition. However, when a novel unregistered face is presented to the system, the result will be misclassification. A more general and challenging open-set face recognition scheme is highly desirable due to its ability in dealing with the unknown persons which are not enrolled before. We observed that there is a shortage in survey papers that explore the research endeavors in open-set face recognition. In this paper, we present a literature survey of the open-set face recognition approaches that have been introduced for real-world scenarios focusing on video surveillance applications. Moreover, we discuss the current difficulties and suggest the promising directions for future research. The paper also describes the evaluation metrics and available benchmarking face video surveillance databases.
تاريخ النشر
24/03/2020
الناشر
Springer
رقم المجلد
632
رقم العدد
ISSN/ISBN
18761100 / 9789811523168
رابط DOI
https://doi.org/10.1007/978-981-15-2317-5_36
الصفحات
425-436
الكلمات المفتاحية
Open-set face recognition, Video surveillance, Identity of Interest (IoI)
وسيم ناهي ابراهيم | Wasseem N. Ibrahem Al-Obaydy | 280