Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Sep 15, 2015

• Create README.md

  nlw0

  committed
  c52a2ca

  Copy full SHA for c52a2ca

Commits on Sep 13, 2011

• Trying to generate some dewarping samples

  

  Nicolau Werneck

  committed
  d248af7

  Copy full SHA for d248af7

Commits on Sep 12, 2011

• Bug squished!

  

  Nicolau Werneck

  committed
  ec05ada

  Copy full SHA for ec05ada

• still hunting ofr the bug

  

  Nicolau Werneck

  committed
  fca284b

  Copy full SHA for fca284b

• still trying to find the bug

  

  Nicolau Werneck

  committed
  60a0575

  Copy full SHA for 60a0575

• Already fitting the model to kinect data. But there is some stupid bug messing things up when Nk != Nl... :(

  

  Nicolau Werneck

  committed
  69247a6

  Copy full SHA for 69247a6

Commits on Sep 11, 2011

• Starting to put all insude a class

  

  Nicolau Werneck

  committed
  72de29a

  Copy full SHA for 72de29a

Commits on Sep 10, 2011

• Regularization tests looking good. Have to put all that into a class now...

  

  Nicolau Werneck

  committed
  792f2da

  Copy full SHA for 792f2da

Commits on Sep 9, 2011

• Calculation of second derivative matrices, and the Laplacean matrix, used for regularization

  

  Nicolau Werneck

  committed
  7a827d6

  Copy full SHA for 7a827d6

• Created a new test, with points sampled fomr an allipsoid. Some results look quite nasty... Lets see if regularization fizes that.

  

  Nicolau Werneck

  committed
  fca58fd

  Copy full SHA for fca58fd

Commits on Sep 8, 2011

• noise

  

  Nicolau Werneck

  committed
  a82f781

  Copy full SHA for a82f781

• Selecting nearest points to model with kdtree, repeatedly. convergence seems ugly, models gets cumpled up in strong curvatures. probably needs regularization...

  

  Nicolau Werneck

  committed
  74eac84

  Copy full SHA for 74eac84

Commits on Sep 7, 2011

• Now we need to implement a procedure that gets the point cloud (q_data) and produces q from the current p0...

  

  Nicolau Werneck

  committed
  eb19059

  Copy full SHA for eb19059

• Playing around with the SVD or matrices U and V.

  

  Nicolau Werneck

  committed
  921b1de

  Copy full SHA for 921b1de

• Mostly nothing. Getting started to create lagrange multipliers based optimziation class. Noise...

  

  Nicolau Werneck

  committed
  9ed4da4

  Copy full SHA for 9ed4da4

Commits on Sep 4, 2011

• testing convergence from less simple conditions... not looking good

  

  Nicolau Werneck

  committed
  4cdf1d7

  Copy full SHA for 4cdf1d7

Commits on Sep 2, 2011

• Cleaning up the code, getting ready for packing the procedures for real life usage.

  

  Nicolau Werneck

  committed
  adf3115

  Copy full SHA for adf3115

Commits on Sep 1, 2011

• Optimization seems to be giving great results. This is just a commit before some radical cleaning and refactoring of the code...

  

  Nicolau Werneck

  committed
  6d62b5e

  Copy full SHA for 6d62b5e

• Optimization with Jacobian calculation seems to be working.

  

  Nicolau Werneck

  committed
  d15943f

  Copy full SHA for d15943f

Commits on Aug 31, 2011

• Jacobian calculation draft is up, but results look terrible.

  

  Nicolau Werneck

  committed
  e812003

  Copy full SHA for e812003

• trying to figure out the Jacobian calculation...

  

  Nicolau Werneck

  committed
  1bfe522

  Copy full SHA for 1bfe522

• ...and Shigeru Ando's consistent filter coefficients.

  

  Nicolau Werneck

  committed
  defb530

  Copy full SHA for defb530

• Including Sacharr filter coefficients just for kicks

  

  Nicolau Werneck

  committed
  608191d

  Copy full SHA for 608191d

Commits on Aug 30, 2011

• Optimizaiton with Lagrange multipliers starting to look good

  

  Nicolau Werneck

  committed
  835f9c7

  Copy full SHA for 835f9c7

• Optimizaiton with Lagrange multipliers starting to look good

  

  Nicolau Werneck

  committed
  409bc86

  Copy full SHA for 409bc86

• optimizaiton results with lagrange multipliers starting to loog good

  

  Nicolau Werneck

  committed
  13e03bb

  Copy full SHA for 13e03bb

• plot results and actual values as images.

  

  Nicolau Werneck

  committed
  456aec7

  Copy full SHA for 456aec7

• Optimization runs but results are crazy.

  

  Nicolau Werneck

  committed
  2331956

  Copy full SHA for 2331956

• Started the optimization with lagrange multipliers. Calculation of U and V matrices looks correct.

  

  Nicolau Werneck

  committed
  898f6f7

  Copy full SHA for 898f6f7

Commits on Aug 22, 2011

• testing normal calculation

  

  Nicolau Werneck

  committed
  873114e

  Copy full SHA for 873114e

Commits on Aug 21, 2011

• Tried to calculate the gradient of kinect data to see how it looks. Way too noisy, we need a (much) better estimation technique for local surface norm...

  

  Nicolau Werneck

  committed
  bfd6e83

  Copy full SHA for bfd6e83

• duh, commnted stupid line. finally seeing the results of dewarping with the camera parameters taken from fit_camera

  

  Nicolau Werneck

  committed
  5304ecf

  Copy full SHA for 5304ecf

Commits on Aug 20, 2011

• Trying to apply camera location results to fit_mapping. There is something wrong...

  

  Nicolau Werneck

  committed
  c15878e

  Copy full SHA for c15878e

Commits on Aug 19, 2011

• Optimization running and producing results that dont seem crazy.

  

  Nicolau Werneck

  committed
  63d7678

  Copy full SHA for 63d7678

• starting the optimization code

  

  Nicolau Werneck

  committed
  e6f3899

  Copy full SHA for e6f3899

Pagination

Previous

Next