DeeplabV3Plus.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from matplotlib import pyplot as pylab\n",
    "import cv2\n",
    "import numpy as np\n",
    "import os\n",
    "\n",
    "from model import Deeplabv3\n",
    "\n",
    "deeplab_model = Deeplabv3()\n",
    "\n",
    "# input directory\n",
    "pathIn = \"imgs\"\n",
    "\n",
    "# output directory\n",
    "pathOut = \"output\"\n",
    "\n",
    "# define params\n",
    "trained_img_width = 512.\n",
    "mean_subtraction_value = 127.5\n",
    "\n",
    "# if the output directory does not exist, create it\n",
    "if not os.path.exists(pathOut):\n",
    "    os.makedirs(pathOut)\n",
    "    \n",
    "img = pylab.imread(pathIn + \"/image1.jpg\")\n",
    "\n",
    "# grab shape of the input image\n",
    "w, h, _ = img.shape\n",
    "\n",
    "# resize the input image\n",
    "ratio = trained_img_width / np.max([w, h])\n",
    "resized = cv2.resize(img, (int(ratio*h), int(ratio*w)))\n",
    "\n",
    "# normalize the resized image\n",
    "resized = resized / mean_subtraction_value - 1.\n",
    "\n",
    "# pad array to square image to match training imgs\n",
    "pad_x = int(trained_img_width - resized.shape[0])\n",
    "resized2 = np.pad(resized, ((0, pad_x), (0, 0), (0, 0)), mode='constant')\n",
    "\n",
    "# make prediction\n",
    "res = deeplab_model.predict(np.expand_dims(resized2, 0))\n",
    "labels = np.argmax(res.squeeze(), -1)\n",
    "\n",
    "# plot sample images\n",
    "pylab.figure(figsize=(10,10))\n",
    "pylab.imshow(labels[:-pad_x], cmap='inferno')\n",
    "pylab.axis('off')\n",
    "pylab.colorbar()\n",
    "pylab.show()\n",
    "pylab.savefig(\"segmented.jpg\", bbox_inches='tight', pad_inches=0)\n",
    "pylab.close()"
   ]
  },
  {
   "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": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}