add nbs

1 files changed, 213 insertions, 0 deletions

diff --git a/megapixels/notebooks/datasets/duke_mtmc/duke_mtmc_timestamps.ipynb b/megapixels/notebooks/datasets/duke_mtmc/duke_mtmc_timestamps.ipynb
new file mode 100644
index 00000000..5179cc7c
--- /dev/null
+++ b/megapixels/notebooks/datasets/duke_mtmc/duke_mtmc_timestamps.ipynb
@@ -0,0 +1,213 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Duke MTMC Timestamps\n",
+    "\n",
+    "- use pymediainfo to extract timestamps\n",
+    "- save data to CSV"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 62,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%reload_ext autoreload\n",
+    "%autoreload 2\n",
+    "\n",
+    "import os\n",
+    "from os.path import join\n",
+    "import math\n",
+    "import time\n",
+    "from glob import glob\n",
+    "import datetime\n",
+    "\n",
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "from pathlib import Path\n",
+    "from tqdm import tqdm_notebook as tqdm\n",
+    "from pymediainfo import MediaInfo"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "fp_dir_videos = '/data_store/datasets/people/duke_mtmc/dataset/videos/'\n",
+    "fp_times = '/data_store/datasets/people/duke_mtmc/processed/video_times.csv'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "fps_video = glob(join(fp_dir_videos, '**/*.MTS'), recursive=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def mediainfo(fp_in, raw=False):\n",
+    "  \"\"\"Get media info using pymediainfo\"\"\"\n",
+    "  \n",
+    "  media_info_raw = MediaInfo.parse(fp_in).to_data()\n",
+    "  media_info = {}\n",
+    "\n",
+    "  if raw:\n",
+    "    for d in media_info_raw['tracks']:\n",
+    "      if d['track_type'] == 'Video':\n",
+    "        media_info['video'] = d\n",
+    "      elif d['track_type'] == 'Audio':\n",
+    "        media_info['audio'] = d\n",
+    "  else:\n",
+    "    for d in media_info_raw['tracks']:\n",
+    "      if d['track_type'] == 'Video':\n",
+    "        media_info['video'] = {\n",
+    "          'codec_cc': d.get('codec_cc', ''),\n",
+    "          'duration': int(d.get('duration','')),\n",
+    "          'display_aspect_ratio': float(d.get('display_aspect_ratio', '')),\n",
+    "          'width': int(d['width']),\n",
+    "          'height': int(d['height']),\n",
+    "          'frame_rate': float(d['frame_rate']),\n",
+    "          'frame_count': int(d['frame_count']),\n",
+    "          }\n",
+    "  \n",
+    "  return media_info"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "87\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(len(fps_video))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 63,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def modification_date(fp):\n",
+    "  t = os.path.getmtime(fp)\n",
+    "  return datetime.datetime.fromtimestamp(t)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 89,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "9cb4c25594cc44d9995da82392acca0a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "HBox(children=(IntProgress(value=0, max=87), HTML(value='')))"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Get file timestamp\n",
+    "meta = []\n",
+    "for fp_video in tqdm(fps_video):\n",
+    "  time_start = modification_date(fp_video)\n",
+    "  camera = int(Path(fp_video).parent.name[-1])\n",
+    "  fn = Path(fp_video).name\n",
+    "  m = mediainfo(fp_video)\n",
+    "  m = m.get('video')\n",
+    "  duration = int(m.get('duration'))\n",
+    "  minutes = duration / 1000 / 60\n",
+    "  time_end = time_start + datetime.timedelta(0, duration//1000) # ms to s\n",
+    "  meta.append(\n",
+    "    {\n",
+    "      'fn': fn, \n",
+    "      'camera': camera,\n",
+    "      'time_start': str(time_start),\n",
+    "      'time_end': str(time_end),\n",
+    "      'duration': duration, # ms\n",
+    "      'frame_count': m.get('frame_count'),\n",
+    "      'frame_rate': m.get('frame_rate'),\n",
+    "      'width': m.get('width'),\n",
+    "      'height': m.get('height'),\n",
+    "      'minutes': f'{minutes:.3f}',\n",
+    "    })"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 90,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "df_meta = pd.DataFrame.from_dict(meta)\n",
+    "df_meta.to_csv(fp_times, index=False)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 94,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Total minutes: 888.7956166666667\n"
+     ]
+    }
+   ],
+   "source": [
+    "print('Total minutes:', df_meta['duration'].sum()/1000/60)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "megapixels",
+   "language": "python",
+   "name": "megapixels"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.6.8"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}