digital_video_introduction/frame_difference_vs_motion_estimation_plus_residual.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Frame difference vs motion estimation + residual"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": false,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.image as mpimg\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfwAAAEPCAYAAACnVHakAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nsvkNMzrpBmYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f837f59e828>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# loading the frames\n",
    "frame_0 = mpimg.imread('i/smw_background_ball_1.png')\n",
    "frame_1 = mpimg.imread('i/smw_background_ball_2.png')\n",
    "# showing the frames\n",
    "f, (plt1, plt2) = plt.subplots(1, 2)\n",
    "plt1.set_title('frame_0');plt1.imshow(frame_0,interpolation='nearest');\n",
    "plt2.set_title('frame_1');plt2.imshow(frame_1,interpolation='nearest');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Frame difference"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWEAAAFiCAYAAAAna2l5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKjKEJami/w/82d/97I64sQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f837c47a748>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.imshow(np.subtract(frame_0[:,:,0], frame_1[:,:,0]),interpolation='nearest',cmap=plt.cm.binary);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAVgAAAFdCAYAAABGoXXzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f8360131d68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# the block we want to motion estimate\n",
    "ball = frame_0[27:37,1:11,]\n",
    "\n",
    "block_size = 10\n",
    "plt.imshow(ball,interpolation='nearest');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Motion estimation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWEAAAFiCAYAAAAna2l5AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nP+is+7YYGQqrAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f83600a20b8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# let's pretend we had motion estimation (full scan, reduced scan(lfp)...) \n",
    "# and that the motion vectors were:\n",
    "x = 26 \n",
    "y = 10\n",
    "predicted_frame = np.array(frame_1)\n",
    "\n",
    "# applying the motion compensation\n",
    "predicted_frame[x:(x+block_size),y:(y+block_size),] = ball\n",
    "\n",
    "plt.imshow(predicted_frame,interpolation='nearest');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Frame difference vs motion estimation + residual"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfwAAAEPCAYAAACnVHakAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x7f835a50a5c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f, (plt1, plt2) = plt.subplots(1, 2)\n",
    "plt1.set_title('Frame difference');plt1.imshow(np.subtract(frame_0[:,:,0], frame_1[:,:,0]),interpolation='nearest',cmap=plt.cm.binary);\n",
    "plt2.set_title('Motion estimation');plt2.imshow(np.subtract(predicted_frame[:,:,0], frame_1[:,:,0]),interpolation='nearest',cmap=plt.cm.binary);"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}