eval/visualize_image.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from IPython.display import Image, display"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from path import Path\n",
    "import struct\n",
    "import os\n",
    "import cv2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_dataset_names = [\"chess-seq-01\", \"chess-seq-02\", \"fire-seq-01\", \"fire-seq-02\", \\\n",
    "                      \"office-seq-01\", \"office-seq-03\", \"redkitchen-seq-01\", \"redkitchen-seq-07\"]\n",
    "labels = [\"input\", \"ground truth\", \"cpp\", \"cpp with ptq\", \"fadec\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "org_dir = Path(\"../dev/dataset_converter/7scenes/images_7scenes\")\n",
    "gt_dir = Path(\"../dev/dataset_converter/7scenes/data_7scenes\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_binary_image(file_path):\n",
    "    data = []\n",
    "    with open(file_path, 'rb') as f:\n",
    "        while True:\n",
    "            d = f.read(1)\n",
    "            if len(d) != 1:\n",
    "                break\n",
    "            data.append(struct.unpack('b', d))\n",
    "    image = np.array(data).reshape(3, 256, 320)[::-1].astype('uint8').transpose(1,2,0)\n",
    "    return cv2.resize(image, (96, 64), interpolation=cv2.INTER_NEAREST)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_binary_depth(file_path):\n",
    "    data = []\n",
    "    with open(file_path, 'rb') as f:\n",
    "        while True:\n",
    "            d = f.read(4)\n",
    "            if len(d) != 4:\n",
    "                break\n",
    "            data.append(struct.unpack('f', d))\n",
    "    depth = np.array(data).reshape(64, 96)\n",
    "    return np.round(np.clip(depth * 72, 0, 255)).astype(np.uint8)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_images(test_dataset_name, file_number):\n",
    "    files = sorted((Path('cpp/results_7scenes') / test_dataset_name).files(\"*.bin\"))\n",
    "    for file_idx in range(len(files)):\n",
    "        if files[file_idx][-10:-4] == file_number:\n",
    "            break\n",
    "    gts = np.load(gt_dir / test_dataset_name + \".npz\")['ground_truth'][1:]\n",
    "    fadecs = np.load(Path('fadec/depths') / test_dataset_name + \".npz\")['depths'][:,0,0,:,:]\n",
    "\n",
    "    images = []\n",
    "    images.append(load_binary_image(org_dir / test_dataset_name / file_number))\n",
    "    images.append(np.round(np.clip(gts[file_idx] * 72, 0, 255)).astype(np.uint8))\n",
    "    images.append(load_binary_depth(Path(\"cpp/results_7scenes\") / test_dataset_name / file_number + \".bin\"))\n",
    "    images.append(load_binary_depth(Path(\"cpp_with_ptq/results_7scenes\") / test_dataset_name / file_number + \".bin\"))\n",
    "    images.append(np.round(np.clip(fadecs[file_idx] * 72, 0, 255)).astype(np.uint8))\n",
    "\n",
    "    return images"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def visualize_images(images, test_dataset_name, file_number):\n",
    "    os.makedirs(\"tmp\", exist_ok=True)\n",
    "    file_names = [\"tmp/%s-%s-%s.png\" % (test_dataset_name, file_number, i) for i in range(len(images))]\n",
    "    for i in range(len(images)):\n",
    "        cv2.imwrite(file_names[i], images[i])\n",
    "\n",
    "    for i in range(len(images)):\n",
    "        print(labels[i])\n",
    "        display(Image(file_names[i]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "input\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ground truth\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cpp\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAGAAAABACAAAAADAXy3SAAAH10lEQVRYCV3B3ZIs11VG0TlXZvWPJEugMG/FHWEICGxXH8Lw7gTXXV1dudfHzqzu1pHH8J+fr6E7CZbWblmWKtkFCHeCOCEiIJOAIAhySALp7fp//svjdXQnHdAqq2qpdVlKIQHCFCYVd4iA7AQEQRDZJYyux//1T7WN0Z0Jq0prWZZ1WUtDQtiFSUVFBdnJTkAQQWQKjLi++q/NGKOTBsqyllqXdVlLCYGwCyIKToggOwFBEBGRKfRwufmncOvR6QBaVtWyTlWGQPiNKCoiyE5AEAQREQKkXTf/rbON7s6Eh1qn06JACF8ERVFkkklAEEFEBAKJy/DfR7Ye3Z1mUstlPa2npQTCdwQRRWSSSRAEERG5S1za/+geo0dPAcWy1tPptCwC4TsiiIggk4AgiIiIHEKq2v8c2Xr0GN2JKKXL6em0lEL4IiCICIJMAoKIiAhyCFb888jWPXqMkSAq1unpcS0xfJFJEEFkEhAQQUQRhBgDS/zL6B7dY+qAotTy8PxYyBfZCYggyCQgCCIiIhjuCv/Su9HbNjogotb69HwC+SA7AUEQZBIQBBERkU+x8K99GNs2RiIiWvXw42MFZJKDgIAgyCQgiCCKIB+CFc99GNu2jQRFpDz9+LREZJKdTIIgICCTICIighxiUDx3ehrbbRsJoqC1/PB0iiLITiYBQZBJQEBERASZwkF86Sk9tts2OqCIujw9n6QQBOQgCAgyCQiIiCACEnbR+NKdnsb7bXRARKx6ej65A5lkJwgCAjIJiCAiyCFAFL+NdHd6vN+2ThQRXR6eT4uKyE4mQUAQkElAEBEEgTAFjd9GutMZ77dtJIogVafnh6VUQCaZBARBQCYBQURA7sIk8Vt3pzt9e99GR0GkXJ+e1lJBJtkJCIKA7ARBBAGZwp3xW3enk9zeb1uHAkSrHp8fSkVADgKCgEwyCYgIsjMcgsRv3Z1O2G7XbQQFxPL045MLCsgkk4CA7GQniAjIpwAR/Nbd6cC2XW+jQQSn5YcfqwpEJpkEZJKdHERE/k7A+K27k8axXd+3BgHF8vEPJ1YQAdkJyEEOAoKC/CZMQfzW3Uljb+/XWxMRRF3/8AO1RAE5yCRfBAQRQQgIgbAzvqQ7acz2/nbrgAii/vSzrCLIQX5PQBBRvhcgYHzJSNLIuF5uHRAEKR5/XbMUAvJFIOwEBBER+RICAfElI0lTGdfLrQOCgLL8+pxakJ38TpgEQUER5C4EQgRfMpI0xbi+3jogICj+9MvCIt+RQwggMikoym9CCIgv6U5aGW+XWwcEBJGHX597EQjIQaYQDoIgiIrcBUKIxpd0J0H6crl1QEAQ8Zd/jAUEkIMQCF8ERRT5ECAB8SXdSSjG5XLrgICA2o//9JiC8EE+hC8yiRMIBAgEIr4k3YkyXt9uHRAQUPAffhEhgEzyJRxkEkRBDoEQIr4knY4yXt9uTWQSEM3DHx+ocJBJfk8+KMinACHiS9JpkHG53JrIJCCiv/yCAQQE5HcEOcgXCZAweSadBBmXy625k0mRpz+W7ASZBCIfBASRXfgUCATPpJMg4+311tzJpOjy65MBQeSTfJBJRIEQDgECAc+kkyDj7fLe3MkkTj//VEFE5CA7uZNJlCkQpkAIk+fQSYR+u7wP7mQSp6efV0QR5CCT3MkkooQQIBAIBM+hkwj9dnlvwkFA1Fp+elJURA7yQSZBEDEEAiEQYvAcOmHq6+XahDsBp6rHH6uwRORODnIQRAQTCIQQiMFz6ISpr5drE+4E3C3L86PlhCA7meQgOxGBQAiEEAieQydM/X65DsIHQUpreXhaLAsFZJJJdnIQASGBEAgB4jl0wtS3y9sgfBDEspbl8VRVTkwyySST3ImAIRBCIEA8h07A5Pb6NgifRKxyWdaHpcodIDsBmeRO5BACgRAgnkOHALld3jbCJxFrt5xOVZYKyEEm+SQCQiAQCGHyHDoEyHa5bIRPItZuWU6rZYmA3MlBdoIgBAKBEDCeQ4cA2d5eN8IX0XKpqmVdqyxFkC8yyU4QDBAIEILxHJoEyHh7vRG+iJa1W9e1tERA7mQndyJ3AQKBGM+xk4A93l5vgcidqDVZy2ld3CHInUzyQeRTIASwPcdOCGa7vr6HMMnOqayy6rSuWoog8kFACDLJp0AIGM+xEwKM6+s1hJ1MolWW1npaF3eIGPkgIB/kUwIB4jl2SCD9/nptwkEmp9LS9WFdLEUNcicgIF/kkBCmeI5pmkDfXq+DcCcgWk61PK5VpaAidwLyHUGmhAAxnmOaJNDb69sgfBBQyt3ydKpyAhUDyCSfZCe7EAgYz6nODnp7fRuEDwLi3fL84AFRJAgIyG8EIRAIUzxjOiRk3C7XQTDsZBIP9fzkDhGRSUC+IzsJEAjYvmB3QpJxe7uOgGEnkyhaPD0tpQgiAnKQ78mHQJD4X6G7E9Jju15HuJODiBPL08NaCAgiyJ18Rz6FSfxbOqMTOtt4f98SQD6JuKuHh7UUmUQQkL8j34kV/zsZozP1GNt164B8EsTd8nCqRREQRARkCnfyG0FS+Lc12xidnrbtfXQinwTEqeq0rlUoOxGRuzDJJB9EkIr/s4zRPXq3bbcxCCBhJ4hT1bqupYIgoiCHMMkkH0SCS/zzT2PrKd1ju91GMwmEnaBo1bouZaGAiPJ7spPflN7+HzsiP6hZH32MAAAAAElFTkSuQmCC",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cpp with ptq\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "fadec\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "test_dataset_name = test_dataset_names[2]\n",
    "file_number = '000139'\n",
    "images = get_images(test_dataset_name, file_number)\n",
    "visualize_images(images, test_dataset_name, file_number)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "input\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ground truth\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cpp\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cpp with ptq\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAGAAAABACAAAAADAXy3SAAAGpUlEQVRYCX3B24pcWXYF0LHXztSlVJefklDRuDHGD6Yf/NZgGgoKQf25LSkzItbcPifyrkxpjPGPX9ssTLu2C9qteC7OWkQSEbeCQizj/8Y/37TJdK9pQtvEc0HbRCISEXeW3bDi8nr8z6XpRhE0WmjfiBuh7SIiG/FgOUtUxh89p3InaJpom3hJ0EgkIuJBsEi6xviX6alGNCFe0IhdJCLisbBIVmqMP3p6ohFNBO25ELuICPFELJGVGuOPnh5rm2gi2hOxC3EjQhC7QhBLZKWMP9sTbRNaQiPuxCY2IcomgrhTISyRFTX+7El70DbREZrYxCY2IYjH4qlYIitlfIpNe6SJlmjEJghBCGJXiF0h7sQSWYxPqaCZ7VaLlmjEJoQggvhGeRDCCj2MTzZxo91o0QmNIIQQIZ4pD4JYJIxPZosb7VZHR2iEEEJEEJvyoNwLwpIwPplanLVbLTpi00RECBFxqzwoN2ITFpHxaaKJTdtNTUsjiJZIiIh4pDwo4iwsIuPTnLrtQmPaNE1smkhnIyLCYrhXvhEWIuPTnLrtQjPdaWdNdNLZSCKWzfCgPBKbhcT4y9QaQTMpT0TrdNK5ZdkNj5WzOFs2SY2/TLoRNJNyY7rT0n1MOjlls+yG71kIUeOviW4EzUTZTKazRkd3OsecTjlZHhm+sWxCGH9NdNvEvbKZprO2aaT7eDqeTqdYnhhuLHeCMP6a6LaLG+VsYmo32ibdx9P16ZQsPxJnYXyapm67OCu7aTe1G22XPh6vT6dTlu+LG8H4NM3WzmJXdtMTjaD7eH04nbLs4l7ZxIOwjPFnme1OUM6mxxqx6T5enw6nLPF9cbYYf8ye7pTHpjttExp9PF2fTqcVN+JG2cQTi/EvKRMT5Ruz3QmNdjxdn06nFfFDi5TxT1GzTKbyPaHRpE/HwyErsVt+IGr8t5TLmmVO5QWxa7RN+nQ8HNJh+aFU1PiHVNW8mDXNQij34kbTztKn68OpE8smfmAN479CVV3WZc2pxEsa7Sx9uj4cVrIQTy0M95Yx/lPqbF5czulWeRCbtmukrw+HU8cizpbvGuM/UKrKxeXl5bQr3wjaruV4fTh0sojd8gNj/DuqVHFx+XZOlOdC27RcXx8OK1kEy0uibGKOv6OqFFVvf5rlRXHWOl+uDulYguVOnJVN3Knx98WoKpuqd7+U5+Jeu/p6dViJJSxn8T01/o3UqCq7V7++KWdB2cVjffX56phYhGUXL1tGjb+hhiq7ize/XdrFWTmLe/3l69UxWYhlE9+xMMffRNUoRVy8+vVNIb6nrz5fHU8WYhEvWO7M8TuqhrK7uHj3ixLf1Vefr44ni7DEN5ZhuTfH7yg1lE1dvPvND/XV16vrZAmWuLG8bIzfLaNqUKiLdz9PuzY907Ndfb06dFZsltgt3zPG74tRZdjVxU+/TZu2mZ5qm+OXL4fOis0Ka7B8zxwfMUoNu7r46edpareme+3s+OXqqrOCJZYfmuOjqFE17Oriza/TDx2vP191VrDE8pIoZ3N8FDWqDJu6ePNzTT9yvP56dWzBEstLohA1x8dQo5SBqjfvZplou+lbx+Pnq2PHZonlXtwoohA1xsdQRhUDF6/eXSpPTQ86x69fjh2bJZZ78VgJNcYHy6hRymbUq3cXU/nGbKZd5/j1y7Fjs8RyL56b44Nl1CgKoy5+em2WR6YHTY7XX647Nkss9+K5OT5YRg1lV+PVm9c1le9Kf/1y3bFZYtmk7OKZOT4sRhkozIs3b8tUQnlB+vrL9TE2SyyblE08N8eHxVDDWdX46W2V6U55ELs+/e/1MZZNLJsoxDM1xoeVGmq4UfX2XZXpB64/Xx3bsollE5R4pub4sFJDGc6q3r65LKYHZRNnRY6fr44dC7Fs4kVr1BwfFqMYzqrevb1gulWeyfHrl2NWWGIhXrZGzfGeoQxnpd69vZjulG/M5vj1y/G0hCUL8cSyGXY1x3uGGm4Ub396PT0od0LZ5Pj16nhahBWL+MbCsJvjPabhVrn89fLSppnOyibOihy/Hg6dJSxZ4pk1bNYwx/uxRhlulde/vbZrTGdlE2eVzvXVqU+WsGLZxYvmeI9Rw52Lt79clmibaVeeyPF4OJw6SyxZxEsWw+V4P5ZRBpahXr95fTFL2p3piXY6ng6nzgpLlri1GDbLnXE5Po5eo4YbVfPi1euLqd2bnurT8XQ4dZawxHIrKMSNZc7xfupRw42q+erNZdnEpjyXHE+HU2cJS5ytYdmN5SysURf/D8K5XtMXBfuCAAAAAElFTkSuQmCC",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "fadec\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "test_dataset_name = test_dataset_names[7]\n",
    "file_number = '000268'\n",
    "images = get_images(test_dataset_name, file_number)\n",
    "visualize_images(images, test_dataset_name, file_number)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.8.2 64-bit ('3.8.2')",
   "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": "3.8.2"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "7666809c90f55d26e137126084f57946dd005ea84acac35a876ccdfdf4177a80"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}