Deep learning for face recognition in surveillance videos | |
| Author | Sarmadi, Paul-Darius |
| Call Number | AIT RSPR no.CS-16-03 |
| Subject(s) | Machine learning Surveillance detection Human face recognition (Computer science) |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Computer Science, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Research studies project report ; no. CS-16-03 |
| Abstract | The problem of recognizing a previously identified criminal, hoping to follow him or her through video cameras feeds is a key issue. However, police require a great deal of human resources to perform this task. Automating the face verification process in surveillance video seems to be a feasible solution to this problem. Deep learning algorithms have recently reached particularly high levels of accuracy in automated face verification. For example, recent approaches reached over 98% accuracy on the “Labeled Faces in the Wild” (LFW) database. The goal of this research is to explore the possibility of adapting such methods to the particular conditions and constraints of video surveillance. I performed experiments on a database of surveillance videos acquired in a crowded shopping mall. With the neural network architecture described in this article, an accuracy of 88.04% was reached on the database. |
| Year | 2016 |
| Corresponding Series Added Entry | Asian Institute of Technology. Research studies project report ; no. CS-16-03 |
| Type | Dissertation |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Dailey, Matthew; |
| Examination Committee(s) | Mongkol Ekpanyapong;Manukid Parnichkun; |
| Scholarship Donor(s) | Telecom SudParis, France;Asian Institute of Technology; |
| Degree | Rsearch Studies Project Report (M. Sc.) - Asian Institute of Technology, 2016 |