Fit Eye Model-Copy1.ipynb

{
 "cells": [
  {
   "metadata": {
    "collapsed": false,
    "trusted": true
   },
   "cell_type": "code",
   "source": "%matplotlib inline\nimport numpy as np\nimport menpo.io as mio\nfrom itertools import chain, islice\nfrom functools import partial\nfrom menpo.feature import ndfeature, fast_dsift",
   "execution_count": null,
   "outputs": []
  },
  {
   "metadata": {
    "collapsed": false,
    "trusted": true
   },
   "cell_type": "code",
   "source": "@ndfeature\ndef menpo_dsift(pixels):\n    return fast_dsift(pixels).astype(np.float32)",
   "execution_count": null,
   "outputs": []
  },
  {
   "metadata": {
    "collapsed": false,
    "trusted": true
   },
   "cell_type": "code",
   "source": "from menpofit.fitter import noisy_shape_from_bounding_box\nfrom menpofit.sdm import Newton\n\nbbox_method = partial(noisy_shape_from_bounding_box, noise_percentage=0.001)\nsdm_alg = partial(Newton, alpha=0.01)",
   "execution_count": null,
   "outputs": []
  },
  {
   "metadata": {
    "collapsed": false,
    "trusted": true
   },
   "cell_type": "code",
   "source": "sdm = SupervisedDescentFitter(images, \n                              group='sclera',\n                              bounding_box_group='bbox',\n                              sd_algorithm_cls=sdm_alg,\n                              patch_features=sparse_hog, \n                              scales=(1, 0.5), \n                              iterations=10, \n                              patch_shape=(17, 17),\n                              n_perturbations=20, \n                              perturb_from_bounding_box=bbox_method,\n                              verbose=False, \n                              batch_size=None)",
   "execution_count": null,
   "outputs": []
  },
  {
   "metadata": {
    "collapsed": false,
    "trusted": true
   },
   "cell_type": "code",
   "source": "sdm.increment(images, \n              group='sclera', \n              bounding_box_group='bbox', \n              verbose=True, \n              batch_size=None)",
   "execution_count": null,
   "outputs": []
  },
  {
   "metadata": {
    "collapsed": false,
    "trusted": true
   },
   "cell_type": "code",
   "source": "print(sdm)",
   "execution_count": null,
   "outputs": []
  },
  {
   "metadata": {
    "collapsed": true,
    "trusted": true
   },
   "cell_type": "code",
   "source": "mio.export_pickle(sdm, '/mnt/data/sdm_test.pkl.gz', overwrite=True)",
   "execution_count": null,
   "outputs": []
  },
  {
   "metadata": {
    "collapsed": false,
    "trusted": true
   },
   "cell_type": "code",
   "source": "from menpofit.fitter import align_shape_with_bounding_box\nfrom menpo.transform import AlignmentAffine, AlignmentSimilarity\n\n\ninit = align_shape_with_bounding_box(sdm.reference_shape, \n                                     im.landmarks['bbox'],\n                                     alignment_transform_cls=AlignmentSimilarity)\nleye_result = sdm.fit(im, init, crop_image=1)",
   "execution_count": null,
   "outputs": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "name": "python2",
   "display_name": "Python 2",
   "language": "python"
  },
  "notify_time": "30",
  "language_info": {
   "mimetype": "text/x-python",
   "nbconvert_exporter": "python",
   "name": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.10",
   "file_extension": ".py",
   "codemirror_mode": {
    "version": 2,
    "name": "ipython"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}