Series Intro Detector.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import cv2\n",
    "import glob\n",
    "import pickle\n",
    "import sys\n",
    "from collections import defaultdict, Counter\n",
    "import caffe\n",
    "from sklearn.neighbors import NearestNeighbors\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "files = glob.glob('./*.avi')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "file_a, file_b = files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "extract_from_layer = \"pool5/7x7_s1\"\n",
    "model_def= \"./deploy.prototxt\"\n",
    "pretrained_model=\"./bvlc_googlenet.caffemodel\"\n",
    "\n",
    "net = caffe.Net(model_def, pretrained_model, caffe.TEST)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7199%"
     ]
    }
   ],
   "source": [
    "def read_video(file_name):\n",
    "    vidcap = cv2.VideoCapture(file_name)\n",
    "    final_arr = []\n",
    "    frame_num = 0\n",
    "\n",
    "    while frame_num < round(vidcap.get(cv2.CAP_PROP_FPS)) * 5 * 60:\n",
    "        sys.stdout.write(\"\\r%d%%\" % frame_num)\n",
    "        sys.stdout.flush()\n",
    "        frame_num += 1\n",
    "        \n",
    "        ret, image = vidcap.read()\n",
    "        img =  cv2.resize(image, (224, 224)) \n",
    "        img = np.array(img,dtype=float)\n",
    "        features = np.zeros(1024,)\n",
    "        img[:,:,[0,1,2]] = img[:, :, [2, 1, 0]] # swap channels\n",
    "        img = img.transpose((2, 0, 1)) # re-order dimensions\n",
    "\n",
    "        net.blobs['data'].reshape(1,3,224,224)\n",
    "        net.blobs['data'].data[...] = img\n",
    "        res = net.forward()\n",
    "        \n",
    "        final_arr.append(np.ndarray.flatten(net.blobs['pool5/7x7_s1'].data[0].copy()))\n",
    "    return final_arr\n",
    "\n",
    "videogen_a = read_video(file_a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "7199%"
     ]
    }
   ],
   "source": [
    "videogen_b = read_video(file_b)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('./House MD 202 - Autopsy.avi', './House MD 203 - Humpty Dumpty.avi')\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "61440"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print (file_a, file_b)\n",
    "import sys\n",
    "sys.getsizeof(videogen_a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "outfile = open('videogen_a.pkl', 'wb')\n",
    "pickle.dump([videogen_a, videogen_b],outfile)\n",
    "outfile.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "infile = open('videogen_a.pkl', 'rb')\n",
    "videogen_a, videogen_b = pickle.load(infile)\n",
    "infile.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "nbrs = NearestNeighbors(n_neighbors=2, algorithm='ball_tree').fit(videogen_a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "distance, indices = nbrs.kneighbors(videogen_b)\n",
    "temp = [i/24 for i in indices]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 30, 156])"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "dp = [1 for i in range(len(temp))]\n",
    "l = [min(i) for ind, i in enumerate(temp)]\n",
    "from scipy.signal import savgol_filter\n",
    "l = savgol_filter(l, 51, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(1, len(l)):\n",
    "    if abs(l[i] - l[i-1]) < 2:\n",
    "        dp[i] = dp[i-1] + 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "161 128\n"
     ]
    }
   ],
   "source": [
    "import operator\n",
    "index, value = max(enumerate(dp), key=operator.itemgetter(1))\n",
    "print index/24, (index-value)/24"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 432x288 with 0 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(l)\n",
    "plt.show()\n",
    "plt.savefig('temp.png')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "house_intro = np.array(videogen_b[index-value:index])\n",
    "house_intro_len = value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}