============================================================================

algorithms.tar contains:

o algorithms
o associated Mex files
o sample code for running each algorithm
o sample data file 
  data format: first column -- x-coordinates, second column -- y-coordinates
	       third column -- flow at x-axis, fourth column -- flow at y-axis.
o the code for generating sample data file.
o other-code: supporting code

============================================================================

simulations.tar contains:

o Simulation for the linear algorithms (Kanatani-A, Kanatani-B, and
  Jepson-Heeger's linear subspace algorithm): sim_linear.m
o Simulation for the nonlinear algorithms (Jepson-Heeger's bilinear
  algorithm, and Tomasi's algorithm): sim_nonlinear.m
o Simulation for Prazdny's rotation-first algorithm: sim_Prazdny.m
o Convergence test for the nonlinear algorithms: sim_convergence.m

Simulations can be performed by varying a large number of parameters
including: noise level, depth range, depth structure, field of view,
translation direction, translational speed, rotation axis, and the
number of sample velocity vectors.

============================================================================

analysis.tar contains:

o Analysis of simulation data: sim_data_analysis.m
0 Analysis of data for convergence test: conv_data_analysis.c
  To compile: cc -g -o conv_data_analysis conv_data_analysis -lm
  The execution of the code gives the following two files:
       [filename]_convergence: Translation error (degrees) vs
                               Percentage of searches
       [filename]_local_minima: Number of local minima vs 
                                Percentage of trials