Name
Abstract | ||
|---|---|---|
Attendance marking is a burdensome and time consuming daily task for a school staff especially the number of students in the classroom are many. Thus, it becomes attractive if this manual marking process can be automated through a real-time facial recognition system [1]. Although facial recognition works well under constrained environment, identifying each individual student in a classroom environment can be very challenging especially the students are in an uncooperative manner. Conventional frame-based accuracy metrics cannot reflect the true outcome of the student attendance as it varies drastically over frames, due to the large variations of scales, poses and occlusions in the actual classroom scenarios. Here, we propose the new accuracy methodology for attendance marking based on statistical multiframes approach. Combined with the sliding window filtering, the system is able to reduce the impacts due to false positive and increasing the overall confidence level of the attendance marking after a convergence time. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1109/TAAI48200.2019.8959936 | 2019 International Conference on Technologies and Applications of Artificial Intelligence (TAAI) |
Keywords | DocType | ISSN |
Smart Campus,Accuracy Metrics,Face Detection (FD),Face Recognition (FR),Deep-Learning (DL),Inter-pupillary Distance (IPD),Attendance Marking,Sliding Window Filtering | Conference | 2376-6816 |
ISBN | Citations | PageRank |
978-1-7281-4667-6 | 0 | 0.34 |
References | Authors | |
4 | 7 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Kuan Heng Lee | 1 | 0 | 0.34 |
| Sanjay V. Addicam | 2 | 0 | 0.34 |
| Ilya Krylov | 3 | 0 | 0.34 |
| Sergei Nosov | 4 | 0 | 0.34 |
| Mee Sim Lai | 5 | 0 | 0.34 |
| Zhan Qiang Lee | 6 | 0 | 0.34 |
| Chung Shien Chai | 7 | 0 | 0.34 |