{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Testing Data\n",
    "\n",
    "blah ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline \n",
    "from matplotlib import pyplot as plt\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import namedtuple\n",
    "import time\n",
    "import pickle\n",
    "import cv2\n",
    "\n",
    "Data_ts = namedtuple('Data_ts', 'data timestamp')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Formats\n",
    "\n",
    "Data was saved to a `dict` with each key being a `list` full of `namedtuples`. The `Data_ts` hold the data plus a time stamp since the start of the data.\n",
    "\n",
    "- `imu`: accel, mag, gyros\n",
    "- `camera`: image bytes\n",
    "- `lidar`: array of (angle, range,) for each point"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "total 615904\r\n",
      "drwxr-xr-x  11 kevin  staff   352B Oct 13 21:50 \u001b[34m.\u001b[m\u001b[m/\r\n",
      "drwxr-xr-x  15 kevin  staff   480B Oct 13 20:31 \u001b[34m..\u001b[m\u001b[m/\r\n",
      "drwxr-xr-x   2 kevin  staff    64B Oct 13 21:50 \u001b[34m.ipynb_checkpoints\u001b[m\u001b[m/\r\n",
      "-rw-r--r--   1 kevin  staff   4.1M Oct 13 20:29 cam.jpg\r\n",
      "-rwxr-xr-x   1 kevin  staff   274K Oct 13 20:29 \u001b[31mdata-work.ipynb\u001b[m\u001b[m*\r\n",
      "-rwx------   1 kevin  staff   293M Oct 13 17:45 \u001b[31mdata.pickle\u001b[m\u001b[m*\r\n",
      "-rwxr-xr-x   1 kevin  staff   1.9K Oct 13 20:29 \u001b[31mget_data.py\u001b[m\u001b[m*\r\n",
      "-rw-r--r--   1 kevin  staff   333K Oct 13 20:29 img.png\r\n",
      "-rwxr-xr-x   1 kevin  staff   1.1K Oct 13 21:38 \u001b[31mtest_ahrs.py\u001b[m\u001b[m*\r\n",
      "-rw-r--r--   1 kevin  staff   936B Oct 13 21:41 test_ins.py\r\n",
      "-rwxr-xr-x   1 kevin  staff   768B Oct 13 21:40 \u001b[31mtest_vo.py\u001b[m\u001b[m*\r\n"
     ]
    }
   ],
   "source": [
    "%ls -alh"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pickle.load( open( \"data.pickle\", \"rb\" ) )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# data['imu'][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# accel = [x[0][0] for x in data['imu']]\n",
    "# mags = [x[0][1] for x in data['imu']]\n",
    "# gyros = [x[0][2] for x in data['imu']]\n",
    "# imutime = [x[1] for x in data['imu']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# plt.figure()\n",
    "# plt.plot(accel);\n",
    "# plt.grid(True);\n",
    "# plt.title(\"Acceration [g's]\")\n",
    "\n",
    "# plt.figure()\n",
    "# plt.plot(mags);\n",
    "# plt.grid(True);\n",
    "# plt.title(\"Magnetometer [uT's']\")\n",
    "\n",
    "# plt.figure()\n",
    "# plt.plot(gyros);\n",
    "# plt.grid(True);\n",
    "# plt.title(\"Gyroscopes [deg/sec]\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "imgs = [np.frombuffer(x[0], dtype=np.uint8).reshape((480,640)) for x in data['camera']]\n",
    "itime = [x[1] for x in data['camera']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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"
    }
   ],
   "source": [
    "plt.imshow(imgs[0], cmap='gray');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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"
    }
   ],
   "source": [
    "plt.imshow(imgs[900], cmap='gray');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "def featureDetection():\n",
    "    thresh = dict(threshold=25, nonmaxSuppression=True);\n",
    "    fast = cv2.FastFeatureDetector_create(**thresh)\n",
    "    return fast"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "def getAbsoluteScale(f0, f1):\n",
    "      x_pre, y_pre, z_pre = f0\n",
    "      x    , y    , z     = f1\n",
    "      scale = np.sqrt((x-x_pre)**2 + (y-y_pre)**2 + (z-z_pre)**2)\n",
    "      return x, y, z, scale\n",
    "\n",
    "def featureTracking(img_1, img_2, p1):\n",
    "\n",
    "    lk_params = dict( winSize  = (21,21),\n",
    "                      maxLevel = 3,\n",
    "                      criteria = (cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_COUNT, 30, 0.01))\n",
    "\n",
    "    p2, st, err = cv2.calcOpticalFlowPyrLK(img_1, img_2, p1, None, **lk_params)\n",
    "    st = st.reshape(st.shape[0])\n",
    "    ##find good one\n",
    "    p1 = p1[st==1]\n",
    "    p2 = p2[st==1]\n",
    "\n",
    "    return p1,p2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "def run(images):\n",
    "    #initialization\n",
    "#     ground_truth =getTruePose()\n",
    "\n",
    "    img_1 = images[0]\n",
    "    img_2 = images[0]\n",
    "\n",
    "    if len(img_1) == 3:\n",
    "        gray_1 = cv2.cvtColor(img_1, cv2.COLOR_BGR2GRAY)\n",
    "        gray_2 = cv2.cvtColor(img_2, cv2.COLOR_BGR2GRAY)\n",
    "    else:\n",
    "        gray_1 = img_1\n",
    "        gray_2 = img_2\n",
    "\n",
    "    #find the detector\n",
    "    detector = featureDetection()\n",
    "    kp1      = detector.detect(img_1)\n",
    "    p1       = np.array([kp.pt for kp in kp1],dtype='float32')\n",
    "    p1, p2   = featureTracking(gray_1, gray_2, p1)\n",
    "\n",
    "    #Camera parameters\n",
    "    fc = 718.8560\n",
    "    pp = (640/2, 480/2)\n",
    "#     K  = getK()\n",
    "\n",
    "    E, mask = cv2.findEssentialMat(p2, p1, fc, pp, cv2.RANSAC,0.999,1.0);\n",
    "    _, R, t, mask = cv2.recoverPose(E, p2, p1,focal=fc, pp = pp);\n",
    "\n",
    "    #initialize some parameters\n",
    "    MAX_FRAME     = 500\n",
    "    MIN_NUM_FEAT  = 150\n",
    "\n",
    "    preFeature = p2\n",
    "    preImage   = gray_2\n",
    "\n",
    "    R_f = R\n",
    "    t_f = t\n",
    "\n",
    "    maxError = 0\n",
    "    ret_pos = []\n",
    "    \n",
    "    for numFrame in range(2, MAX_FRAME):\n",
    "\n",
    "        if numFrame % 20 == 0:\n",
    "            print(numFrame)\n",
    "\n",
    "        if (len(preFeature) < MIN_NUM_FEAT):\n",
    "            feature   = detector.detect(preImage)\n",
    "            preFeature = np.array([ele.pt for ele in feature],dtype='float32')\n",
    "            print(\">> features found: \", len(preFeature))\n",
    "            if len(preFeature) < MIN_NUM_FEAT:\n",
    "                continue\n",
    "\n",
    "        curImage_c = images[numFrame]\n",
    "\n",
    "        if len(curImage_c) == 3:\n",
    "              curImage = cv2.cvtColor(currImage_c, cv2.COLOR_BGR2GRAY)\n",
    "        else:\n",
    "              curImage = curImage_c\n",
    "\n",
    "        kp1 = detector.detect(curImage);\n",
    "        preFeature, curFeature = featureTracking(preImage, curImage, preFeature)\n",
    "        E, mask = cv2.findEssentialMat(curFeature, preFeature, fc, pp, cv2.RANSAC,0.999,1.0);\n",
    "        \n",
    "#         print(E)\n",
    "        \n",
    "        _, R, t, mask = cv2.recoverPose(E, curFeature, preFeature, focal=fc, pp = pp);\n",
    "\n",
    "#         truth_x, truth_y, truth_z, absolute_scale = getAbsoluteScale(\n",
    "#             ground_truth[numFrame-1], ground_truth[numFrame])\n",
    "        \n",
    "#         if numFrame % 20 == 0:\n",
    "#             print('scale', absolute_scale)\n",
    "\n",
    "        absolute_scale = 1.0\n",
    "            \n",
    "        if absolute_scale > 0.1:\n",
    "            t_f = t_f + absolute_scale*R_f.dot(t)\n",
    "            R_f = R.dot(R_f)\n",
    "        else:\n",
    "            print(\"crap ... bad scale:\", absolute_scale)\n",
    "\n",
    "        preImage = curImage\n",
    "        preFeature = curFeature\n",
    "\n",
    "#         ret_pos.append((t_f[0], t_f[2],))\n",
    "        ret_pos.append(t_f)\n",
    "\n",
    "    return ret_pos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "20\n",
      ">> features found:  638\n",
      ">> features found:  87\n",
      "40\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "60\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "80\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "100\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "120\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "140\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "160\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "180\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "200\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "220\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "240\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "260\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "280\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "300\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "320\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "340\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "360\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "380\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "400\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "420\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "440\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "460\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      "480\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n",
      ">> features found:  87\n"
     ]
    }
   ],
   "source": [
    "pos = run(imgs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "def getK():\n",
    "    return   np.array([[7.188560000000e+02, 0, 6.071928000000e+02],\n",
    "              [0, 7.188560000000e+02, 1.852157000000e+02],\n",
    "              [0, 0, 1]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "x and y can be no greater than 2-D, but have shapes (37,) and (37, 3, 1)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-56-049bf664f8b3>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpos\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/matplotlib/pyplot.py\u001b[0m in \u001b[0;36mplot\u001b[0;34m(scalex, scaley, data, *args, **kwargs)\u001b[0m\n\u001b[1;32m   2747\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mscalex\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mscaley\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mTrue\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2748\u001b[0m     return gca().plot(\n\u001b[0;32m-> 2749\u001b[0;31m         *args, scalex=scalex, scaley=scaley, data=data, **kwargs)\n\u001b[0m\u001b[1;32m   2750\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2751\u001b[0m \u001b[0;31m# Autogenerated by boilerplate.py.  Do not edit as changes will be lost.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/matplotlib/__init__.py\u001b[0m in \u001b[0;36minner\u001b[0;34m(ax, data, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1783\u001b[0m                         \u001b[0;34m\"the Matplotlib list!)\"\u001b[0m \u001b[0;34m%\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mlabel_namer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__name__\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1784\u001b[0m                         RuntimeWarning, stacklevel=2)\n\u001b[0;32m-> 1785\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mfunc\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0max\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1786\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1787\u001b[0m         inner.__doc__ = _add_data_doc(inner.__doc__,\n",
      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/matplotlib/axes/_axes.py\u001b[0m in \u001b[0;36mplot\u001b[0;34m(self, scalex, scaley, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1602\u001b[0m         \u001b[0mkwargs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcbook\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mnormalize_kwargs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmlines\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mLine2D\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_alias_map\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1603\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1604\u001b[0;31m         \u001b[0;32mfor\u001b[0m \u001b[0mline\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_get_lines\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1605\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd_line\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mline\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1606\u001b[0m             \u001b[0mlines\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mline\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/matplotlib/axes/_base.py\u001b[0m in \u001b[0;36m_grab_next_args\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    391\u001b[0m                 \u001b[0mthis\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    392\u001b[0m                 \u001b[0margs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 393\u001b[0;31m             \u001b[0;32myield\u001b[0m \u001b[0;32mfrom\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_plot_args\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mthis\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    394\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    395\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/matplotlib/axes/_base.py\u001b[0m in \u001b[0;36m_plot_args\u001b[0;34m(self, tup, kwargs)\u001b[0m\n\u001b[1;32m    368\u001b[0m             \u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mindex_of\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtup\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    369\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 370\u001b[0;31m         \u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_xy_from_xy\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    371\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    372\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcommand\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'plot'\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m/usr/local/lib/python3.7/site-packages/matplotlib/axes/_base.py\u001b[0m in \u001b[0;36m_xy_from_xy\u001b[0;34m(self, x, y)\u001b[0m\n\u001b[1;32m    232\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mndim\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m2\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0my\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mndim\u001b[0m \u001b[0;34m>\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    233\u001b[0m             raise ValueError(\"x and y can be no greater than 2-D, but have \"\n\u001b[0;32m--> 234\u001b[0;31m                              \"shapes {} and {}\".format(x.shape, y.shape))\n\u001b[0m\u001b[1;32m    235\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    236\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mx\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mndim\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mValueError\u001b[0m: x and y can be no greater than 2-D, but have shapes (37,) and (37, 3, 1)"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(pos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[array([[-0.40693868],\n",
       "        [-0.7098152 ],\n",
       "        [ 0.69409058]]), array([[ 0.57504107],\n",
       "        [-0.68305969],\n",
       "        [ 0.50700797]]), array([[ 1.00156704],\n",
       "        [-0.52120521],\n",
       "        [ 1.39688365]]), array([[ 1.94562841],\n",
       "        [-0.50226384],\n",
       "        [ 1.06765826]]), array([[ 0.95371139],\n",
       "        [-0.53147534],\n",
       "        [ 1.19113825]]), array([[ 1.89245295],\n",
       "        [-0.23279983],\n",
       "        [ 1.01921616]]), array([[ 2.70181388],\n",
       "        [-0.26300359],\n",
       "        [ 1.60575058]]), array([[ 1.72023893],\n",
       "        [-0.4457358 ],\n",
       "        [ 1.66160363]]), array([[ 0.7392576 ],\n",
       "        [-0.55217003],\n",
       "        [ 1.82392241]]), array([[ 0.60492747],\n",
       "        [-0.73916627],\n",
       "        [ 2.7970554 ]]), array([[ 0.64647299],\n",
       "        [-0.76910904],\n",
       "        [ 3.79574324]]), array([[ 0.71596175],\n",
       "        [-1.07172295],\n",
       "        [ 4.74631998]]), array([[ 0.65088957],\n",
       "        [-0.57350075],\n",
       "        [ 3.88171588]]), array([[-0.34342886],\n",
       "        [-0.59276203],\n",
       "        [ 3.98640531]]), array([[-1.33294179],\n",
       "        [-0.73616096],\n",
       "        [ 4.00375199]]), array([[-1.88872396],\n",
       "        [-0.22067053],\n",
       "        [ 4.65596038]]), array([[-2.14581001],\n",
       "        [ 0.24021071],\n",
       "        [ 5.50536914]]), array([[-1.26765447],\n",
       "        [ 0.59585454],\n",
       "        [ 5.18543116]]), array([[-0.3892686 ],\n",
       "        [ 0.92426476],\n",
       "        [ 4.83817768]]), array([[0.33484784],\n",
       "        [1.4199956 ],\n",
       "        [5.31766316]]), array([[-0.45980764],\n",
       "        [ 0.92931562],\n",
       "        [ 5.67509262]]), array([[0.09077144],\n",
       "        [0.26496665],\n",
       "        [5.16961945]]), array([[0.75041384],\n",
       "        [0.95059502],\n",
       "        [4.86174667]]), array([[1.31170831],\n",
       "        [1.72660502],\n",
       "        [4.5740715 ]]), array([[0.63901785],\n",
       "        [1.01041252],\n",
       "        [4.75996353]]), array([[1.4359498 ],\n",
       "        [0.48868169],\n",
       "        [4.45550268]]), array([[0.68592034],\n",
       "        [0.79737185],\n",
       "        [5.04045242]]), array([[-0.25134448],\n",
       "        [ 1.09312078],\n",
       "        [ 5.2250255 ]]), array([[-0.92224575],\n",
       "        [ 1.75706833],\n",
       "        [ 5.55527562]]), array([[-0.27916358],\n",
       "        [ 1.15836975],\n",
       "        [ 5.07777658]]), array([[0.46727179],\n",
       "        [0.6704677 ],\n",
       "        [4.62524391]]), array([[0.70425332],\n",
       "        [1.50587395],\n",
       "        [5.1211634 ]]), array([[1.64830925],\n",
       "        [1.83323214],\n",
       "        [5.08122577]]), array([[1.5971375 ],\n",
       "        [2.65219239],\n",
       "        [5.6527899 ]]), array([[2.15210346],\n",
       "        [3.42822475],\n",
       "        [5.95243395]]), array([[1.55411176],\n",
       "        [3.9982666 ],\n",
       "        [6.51586824]]), array([[2.42564451],\n",
       "        [4.45731757],\n",
       "        [6.34352302]])]"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [],
   "source": [
    "aa=np.array([1,2,3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 2, 3])"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aa.transpose()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 3])"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aa[::2]"
   ]
  },
  {
   "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": "3.7.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}