======================
==== Introduction ====
======================

This is a Matlab implementation of the handwritten digit recognition
algorithm from:

    Geoffrey E. Hinton, Christopher K. I. Williams, and Michael Revow. Adaptive elastic models for hand-printed character recognition. NIPS 1992.

====================
==== How to Run ====
====================

To run a quick demo, first make sure the requirements are fulfilled
(see below). Then, run the do_classify_digit.m script:

    >> do_classify_digit

This will run the elastic matching algorithm on several MNIST
images, and output the fitting results to the user in a visual
form.

======================
==== Requirements ====
======================

This has been tested on Matlab R2013a. It should work on later 
Matlab versions.

This requires the third-party library CVX, a wonderful convex
optimization suite found here:
    http://cvxr.com/cvx/

Finally, you will need to download the MNIST handwritten digit
dataset:
    http://yann.lecun.com/exdb/mnist/

===============
==== About ====
===============

This work was completed for the CS 269 (Visual Modeling) course
project, taught by Professor Demetri Terzopoulos (Fall 2014).

The group members are:
    Ansuya Ahluwalia (ansuya@cs.ucla.edu)
    Eric Kim (ekim@cs.ucla.edu)
    Nicholas Brett Marcott (bmarcott@ucla.edu)

