Research ->Projects, Publications, Software & Datasets, People Software & Datasets-> Neural Network Eye Tracker (NNET) Neural Network Eye Tracker (NNET) (code, documentation) This distribution contains the code and documentation for Neural Network Eye Tracker that can be run on an unmodified mobile platform. NNET can be used as an educational tool to depict the challenges of eye tracking and, possibly, provide a baseline in terms of what positional accuracy and sampling rate are possible on a mobile platform. In the best case we were able to achieve sampling rate of 5Hz and the positional accuracy of 3.7 degrees of the visual angle, however average performance numbers are lower. Currently our implementation was tested only on iPad 2. NNET is available via App Store for free. In its current for NNET performs calibration procedure, then verification of calibration, which displays frame rate and positional accuracy in pixels. After that NNET gives an opportunity to play a card game where two of the same cards should be selected to achieve victory. More detailed technical description of NNET is available here. Released code and App Store application has "Face Detection" and "Eye Detection" modules disabled to provide fluid user experience at the expense of lowered positional accuracy. It is possible to enable all modules in the code. Download NNET source code here. If you download the software it is assumed that you agree to the copyright agreement at the bottom of the page. The compressed files have been password protected. Please email Dr. Oleg Komogortsev for the password. Kindly indicate your university/industry affiliation and a brief description of how you plan to use the software. Please put "NNET" in the subject line. Acknowledgment: special thanks are expressed to Mr. Corey Holland for the design and implementation of this software. Currently this project is funded in part by the NSF CAREER award #CNS-1250718, in part by the NSF GRFP award #DGE- 11444666, in part by the #60NANB12D234 grant from the National Institute of Standards and funds from Texas State University. In the past this project was funded in part by the grant #60NANB10D213 from the National Institute of Standards. COPYRIGHT NOTICE STARTS HERE----------------------------------------------------------- Copyright
© 2024 The Texas State University The software is distributed under the following license and to be cited in the bibliography as:
C. Holland and O. V. Komogortsev, Eye Tracking on Unmodified Common Tablets: Challenges and Solutions, In Proceedings of ACM Eye Tracking Research & Applications Symposium, Santa Barbara, CA, 2012, pp. 1-4. IN NO EVENT SHALL THE TEXAS STATE UNIVERSITY BE LIABLE TO ANY PARTY FOR DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OF THIS SOFTWARE AND ITS DOCUMENTATION, EVEN IF THE TEXAS STATE UNIVERSITY HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. THE TEXAS STATE UNIVERSITY SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE SOFTWARE PROVIDED ON AN "AS IS" BASIS, AND THE TEXAS STATE UNIVERSITY HAS NO OBLIGATION TO PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS COPYRIGHT NOTICE ENDS HERE---------------------------------------------------------- |
|