-
Notifications
You must be signed in to change notification settings - Fork 3
Home
This project aims at studying correlation techniques for facial recognition. We explore the capabilities of correlation techniques as a non-segmentation alternative for biometric recgnition.
The main objective of the project is to perform subject identification using correlation techniques. In this project we aim at exploring two techniques: MACE filtering and Composite Filtering. Those techniques are based on a simple principle. A database of images is acquired, and a filter is devised in the image frequency domain so that matching is performed via correlation.
The fundamental references of the project are:
-
[2]. Composite versus multichannel binary phase-only filtering. de la Tocnaye, Quemener & Petillot.
-
[3]. A Technique for Optically Convolving Two Functions. C. S. Weaver and J. W. Goodman
-
[4]. Signal detection by complex spatial filtering. A. V. Lugt
-
[5]. Face Verification using Correlation Filters. Kumar et. Al.
-
[6]. MACE Correlation Filter Algorithm for Face Verification ni Surveillance. Omidora et. Al.
-
[8]. New Perspectives in face Correlation Research: A Tutorial. Wang et. Al