\n",
"\n",
" Description: Given a training set of 11,000 64x64 training images of faces, and a test set of 1,233 64x32 training images containing only the left side of a face image, reconstruct the right sides of the test images. This reconstruction will use the linear least squares regression technique.
"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# The libraries that will be used\n",
"\n",
"%matplotlib inline\n",
"import matplotlib.pyplot as plt\n",
"import numpy as np\n",
"import os\n",
"import random \n",
"import scipy.io\n",
"import urllib"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 1
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"
Getting the Data
\n",
"First a new folder for the image data called 'facial_reconstruction' is made in the current working directory . This new directory is made the current working directory and the image data is downloded to it. The image data is in matlab matrix format, so they must be converted to numpy objects.\n"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"os.mkdir(os.getcwd() + '/facial_reconstruction') # make a directory called Facial_Reconstruction\n",
"os.chdir('facial_reconstruction') # set working directory to the direct\n",
"print 'Image data files will be downloaded to {}'.format(os.getcwd()) # show location data is downloading to"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Image data files will be downloaded to /home/neil/Facial-Reconstruction/facial_reconstruction\n"
]
}
],
"prompt_number": 2
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"data_location = \"http://www.sci.ccny.cuny.edu/~szlam/2013-fall-366/\" # url of image data data\n",
"image_stack_names = ['train_data_left', # names of the data matrices\n",
" 'train_data_right',\n",
" 'test_data_left']\n",
"img_stacks = [] # this will hold the downloaded image data as numpy matrices \n",
"for index, name in enumerate(image_stack_names): \n",
" file_name = name + '.mat' # creates filename used to get and name the data\n",
" urllib.urlretrieve(data_location + file_name, file_name) # download the file to the Facial Recognition folder\n",
" print 'File {} of 3 has finished downloading'.format(index+1) # send a message that the image data downloaded successfully\n",
" img_stacks.append(scipy.io.loadmat(name)[name]) # convert matlab data to numpy data and stores it in a list\n",
"print 'The contents of {} are {}'.format(os.getcwd(), os.listdir(os.getcwd())) # make sure the data downloaded and is in the folder"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"File 1 of 3 has finished downloading\n",
"File 2 of 3 has finished downloading"
]
},
{
"output_type": "stream",
"stream": "stdout",
"text": [
"\n",
"File 3 of 3 has finished downloading"
]
},
{
"output_type": "stream",
"stream": "stdout",
"text": [
"\n",
"The contents of /home/neil/Facial-Reconstruction/facial_reconstruction are ['train_data_left.mat', 'test_data_left.mat', 'train_data_right.mat']\n"
]
}
],
"prompt_number": 3
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"
Reshaping the Data
\n",
"The data will be checked by looking at the first image in the training and testing stacks. The data will be confirmed to be held in a 3-dimensional stack of images, converted to a 2-dimensional stack where each row contains an image in vector form, and checked again. "
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plt.gray()\n",
"plt.imshow(np.hstack((\n",
" img_stacks[0][:,:,0], \n",
" img_stacks[1][:,:,0],\n",
" ))\n",
" )\n",
"\n",
"print 'Viewing first complete image from the 2 stacks of {} {}x{} training images'.format(\n",
" img_stacks[0].shape[2],\n",
" img_stacks[0].shape[0],\n",
" img_stacks[0].shape[1],\n",
" )"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Viewing first complete image from the 2 stacks of 12000 64x32 training images\n"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAP0AAAD+CAYAAADxoQNSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
"text": [
""
]
}
],
"prompt_number": 4
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plt.imshow(img_stacks[2][:,:,0])\n",
"\n",
"print 'Viewing first image from the stack of {} {}x{} test images'.format(img_stacks[2].shape[2],\n",
" img_stacks[2].shape[0],\n",
" img_stacks[2].shape[1]) "
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Viewing first image from the stack of 1233 64x32 test images\n"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAI0AAAD+CAYAAAAH6IbKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nG2F467+HAAAAAElFTkSuQmCC\n",
"text": [
""
]
}
],
"prompt_number": 5
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# To train the filter I need to take the 3-d stacks of images and turn them into 2-d matrices \n",
"# Each image will be turned into a vector and stored as a row in the corresponding matrix\n",
"\n",
"def convert_to_rows(stack):\n",
" '''Takes a stack of images and creates a numpy matrix where every row is an image'''\n",
" size = stack.shape[2] # get number of pictures in stacks\n",
" images = np.empty((size, 2048)) # create empty matrix to hold image vectors\n",
" for i in range(size): # loop through each image in the stack\n",
" images[i] = stack[:,:,i].reshape(2048,) # reshape image to a vector and store it in the matrix\n",
" return images\n",
"\n",
"img_rows = [] # this will hold our converted stacks of images \n",
"for stack in img_stacks: # loop through the 3 stacks of images\n",
" img_rows.append(convert_to_rows(stack)) # convert the stacked images into vector images "
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 6
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Let's make sure that the stacks of images were converted correctly and that each row is indeed an image\n",
"\n",
"plt.gray()\n",
"plt.imshow(np.hstack((\n",
" img_rows[0][0].reshape(64, 32), \n",
" img_rows[1][0].reshape(64, 32),\n",
" ))\n",
" )\n",
"\n",
"print 'Viewing first rows/images from the 2 {}x{} training matrices'.format(img_rows[0].shape[0],\n",
" img_rows[0].shape[1],)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Viewing first rows/images from the 2 12000x2048 training matrices\n"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAP0AAAD+CAYAAADxoQNSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
"text": [
""
]
}
],
"prompt_number": 7
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"# Now to check that the first row/image in the test matrix\n",
"\n",
"plt.imshow(img_rows[2][0].reshape(64, 32))\n",
"\n",
"print 'Viewing first row/image from the {}x{} test matrix'.format(img_rows[0].shape[0],\n",
" img_rows[0].shape[1])"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"Viewing first row/image from the 12000x2048 test matrix\n"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAI0AAAD+CAYAAAAH6IbKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nG2F467+HAAAAAElFTkSuQmCC\n",
"text": [
""
]
}
],
"prompt_number": 8
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"
Regression and Reconstruction
\n",
"Now a 2,048 square weight matrix will be found that minimizes the least squares error between L\\*w and R, \n",
"where L and R are 12,000 x 2,0148 matrices containing the left and right sides of the training data images respectively. Then the 1,233x2,048 test matrix is run through w to reconstruct the right side of the test images so that test * w = reconstructed. Finding this reconstructed matrix completes the goal of this project. "
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"left, right, test = img_rows # unpack our images matrices to convenient names \n",
"weights = np.linalg.lstsq(left, right)[0] # use least squares regression to find weight matrix that reconstructs partial facial images\n",
"reconstructed = np.dot(test, weights) # then reconstruct the other half of the test set images "
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 9
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"