View Algorithm

Accuracy 0.5: 98.54
Global Accuracy: 90.07
Valid images (%): 99.92
Authors: Evangelos Skodras and Nikolaos Fakotakis
Filiation: University of Patras, Patras, Greece
Link to group/authors: http://www.wcl.ece.upatras.gr/ai/people/skodras-evangelos
Algorithm name: Eye Loc
Reference if published: Evangelos Skodras and Nikolaos Fakotakis, "An Accurate Eye Center Localization Method for Low Resolution Color Imagery", International Conference on Tools with Artificial Intelligence (ICTAI 2012), Athens, Greece, November 7-9, 2012
Development platform: Matlab
Execution time (indicating execution platform):Not Specified
Memory requisites: Not Specified
Training requisites: -
Link to source: https://dl.dropboxusercontent.com/u/11073387/Eye_Localization_Code.zip
Abstract (5000 chars): The localization of eye centers and tracking of gaze constitutes an integral component of many human - computer interaction applications. A number of constraints including intrusiveness, mobility, robustness and high-price of eye tracking systems have hindered the way of the integration of eye trackers in everyday applications. Several 'passive' systems based on a single camera have been lately proposed in the literature, exhibiting though subordinate precision compared to the commercial, hardware-based eye tracking devices. We introduce an automatic,non intrusive method for precise eye center localization in low resolution images, acquired from single low-cost cameras. To this end, the proposed system uses color information to derive eye maps that emphasize the iris area and a radial symmetry transform which operates both on the original eye images and the eye maps. Experimental results demonstrate great accuracy in challenging cases and resilience to pose and illumination variations.