## [Vxl-users] vnl least squares minimization

 [Vxl-users] vnl least squares minimization From: Nick Hurlburt - 2002-03-07 18:06:15 ```I wrote a bundle adjustment routine to solve an N-view reconstruction problem using vnl_levenberg_marquardt. My vnl_least_squares subclass takes a vector that is the serialization of the projection matrices and returns the vector whose elements are geometric distances dist(PX, x), where X and x are world and image points that are kept unchanged. There are about 20*N image points and N is only 3 or 4 right now. I am not supplying a gradient. The initial guess is reasonably good. Most of the time, the solver responds with something like: vnl_levenberg_marquardt: too many iterations vnl_levenberg_marquardt: 390 iterations, 14403 evaluations, 177 residuals. RMS error start/end 1.8665/0.422231 and then fails. From intermediate measurements I know the solution is converging, albeit slowly and not strictly monotonically, but the solver cuts off and throws away the incomplete but improved results. Can I change the convergence criteria or at least recover the last parameter vector? Also, if you see a way by which I can improve the general algorithm I've described, I'd be interested. On the rare times it does terminate, the error from external reprojection tests isn't always better, sometimes a little worse. __________________________________________________ Do You Yahoo!? Try FREE Yahoo! Mail - the world's greatest free email! http://mail.yahoo.com/ ```

 [Vxl-users] vnl least squares minimization From: Nick Hurlburt - 2002-03-07 18:06:15 ```I wrote a bundle adjustment routine to solve an N-view reconstruction problem using vnl_levenberg_marquardt. My vnl_least_squares subclass takes a vector that is the serialization of the projection matrices and returns the vector whose elements are geometric distances dist(PX, x), where X and x are world and image points that are kept unchanged. There are about 20*N image points and N is only 3 or 4 right now. I am not supplying a gradient. The initial guess is reasonably good. Most of the time, the solver responds with something like: vnl_levenberg_marquardt: too many iterations vnl_levenberg_marquardt: 390 iterations, 14403 evaluations, 177 residuals. RMS error start/end 1.8665/0.422231 and then fails. From intermediate measurements I know the solution is converging, albeit slowly and not strictly monotonically, but the solver cuts off and throws away the incomplete but improved results. Can I change the convergence criteria or at least recover the last parameter vector? Also, if you see a way by which I can improve the general algorithm I've described, I'd be interested. On the rare times it does terminate, the error from external reprojection tests isn't always better, sometimes a little worse. __________________________________________________ Do You Yahoo!? Try FREE Yahoo! Mail - the world's greatest free email! http://mail.yahoo.com/ ```
 RE: [Vxl-users] vnl least squares minimization From: Ian Scott - 2002-03-11 10:57:05 ```Nick, > Can I change the > convergence criteria or at > least recover the last parameter vector? http://www.isbe.man.ac.uk/public_vxl_doc/vxl/vnl/html/class_vnl_nonlinear_mi nimizer.html The base vnl_nonlinear_minimizer has functions such as set_f_tolerance(), and set_max_function_evals(). The call to minimize(x) should have left x with the best value it found. If not, then that is a bug. Ian. ```
 [Vxl-users] documentation From: Nick Hurlburt - 2002-04-29 19:10:51 ```I have some code that I would like to document using doxygen in the same style as the VXL code (with the intention of maybe contributing some of it at some point). The doxygen manual doesn't discuss the "\\:" format that seems to be used. Could someone explain how I can use the same style, or maybe send me a sample config file? Thanks, Nick Hurlburt __________________________________________________ Do You Yahoo!? Yahoo! Health - your guide to health and wellness http://health.yahoo.com ```