{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "# Stereo Images" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "CameraTransform comes with a basic functionality to use stereo rigs. To calibrate a stereo rig, point correspondences in both images are needed. Stereo rigs should have two cameras that have different positions and orientations in space.\n", "\n", "We start with an example of two camera images:" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [], "source": [ "import matplotlib.pyplot as plt\n", "\n", "imA = plt.imread(\"StereoLeft.jpg\")\n", "imB = plt.imread(\"StereoRight.jpg\")" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ ".. figure:: StereoLeft.jpg\n", " :alt: The left stereo image\n", "\n", " The left image of the stereo rig.\n", " \n", ".. figure:: StereoRight.jpg\n", " :alt: The right stereo image\n", "\n", " The right image of the stereo rig." ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Setting up" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ ".. currentmodule:: cameratransform\n", "\n", "We assume that the internal parameters of the camera (lens distortion or projection) have already been obtained. We now start with creating two cameras that share the same projection and create a CameraGroup object of both." ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [], "source": [ "import cameratransform as ct\n", "\n", "cam1 = ct.Camera(ct.RectilinearProjection(focallength_px=3863.64, image=[4608, 2592]))\n", "cam2 = ct.Camera(cam1.projection)\n", "\n", "cam_group = ct.CameraGroup(cam1.projection, (cam1.orientation, cam2.orientation))" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "As a stereo setup cannot determine the length of objects in the scene if no length is provided, we have to define a baseline (distance between the two cameras). If the baseline is not known, it can be set to 1 and later the baseline can be scaled so that an object in the scene with known length has the right dimenstions." ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [], "source": [ "baseline = 0.87" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "Now we define an initial guess of the parameters:" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [], "source": [ "cam1.tilt_deg = 90\n", "cam1.heading_deg = 35\n", "cam1.roll_deg = 0\n", "cam1.pos_x_m = 0\n", "cam1.pos_y_m = 0\n", "cam1.elevation_m = 0\n", "\n", "cam2.tilt_deg = 90\n", "cam2.heading_deg = -18\n", "cam2.roll_deg = 0\n", "cam2.pos_x_m = baseline\n", "cam2.pos_y_m = 0\n", "cam2.elevation_m = 0" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Fitting" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "To fit the relative orientations of the images, we need point correspondences. This is then added as information to the camera group." ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [], "source": [ "import numpy as np\n", "corresponding1 = np.array([[1035.19720133, 1752.47539918], [1191.16681219, 1674.08229634], [1342.64515152, 1595.28089609], [1487.18243488, 1525.87033629], [1623.55377003, 1456.86807389], [1753.39234661, 1391.94878561], [1881.18943613, 1331.92906624], [2002.45376709, 1275.17572617], [2119.63512393, 1220.46387315], [2229.46712739, 1167.79350718], [1132.21312835, 1106.84930585], [ 902.38609626, 1016.70756572], [1594.61961207, 628.61640974], [1794.85813404, 704.30794726], [1791.41760961, 663.70975895], [1956.56278237, 634.80935372], [2121.70795513, 600.40410939], [2288.22933766, 571.50370416], [2439.61241269, 544.66761359], [2583.42633397, 517.14341813], [2736.18561877, 488.93111778], [2876.55901561, 462.7831321 ], [3010.05136359, 437.3232513 ], [3144.23181646, 413.92768515], [3270.84311557, 393.28453856], [3393.32578537, 369.88897242], [2688.73357512, 1033.7464922 ], [2885.97789523, 1144.56386559], [3068.37586866, 1053.89510554], [2870.60132189, 952.09158548], [3237.51817542, 641.90898531], [3288.95016212, 879.98075878], [4105.49922925, 943.60795881], [3965.51938917, 692.28051867], [4037.41700677, 1135.08413333], [3253.6366414 , 1071.87603935], [3769.95210584, 1204.30954628], [4207.84687809, 1265.15387253], [3808.67122255, 1451.37438621], [3587.10046207, 1264.66938952], [3639.1604923 , 1353.23128002]])\n", "corresponding2 = np.array([[ 352.50914077, 801.67993473], [ 460.08683446, 810.68133359], [ 570.95772285, 820.12182507], [ 682.48725018, 829.56231656], [ 797.35908021, 838.80579923], [ 908.18086062, 848.66284585], [1020.80811952, 858.76197628], [1135.19174893, 870.61747721], [1249.57537835, 882.03388552], [1365.71537827, 893.45029383], [ 911.70331034, 537.73870329], [ 956.46082091, 454.7508191 ], [1718.73717286, 432.37206381], [1706.61534708, 518.62351648], [2062.64027012, 573.54205381], [2203.84185338, 600.78475058], [2355.65939432, 631.55257909], [2518.58129093, 665.72644033], [2673.15840747, 699.10556061], [2837.669786 , 733.27942185], [3018.47335419, 768.64539453], [3202.45588621, 807.58770152], [3386.83578872, 844.9405266 ], [3584.72628753, 889.04864983], [3786.59049113, 932.3620321 ], [3991.63365858, 976.47015533], [2525.23601839, 1218.44516641], [2474.07408151, 1366.28811638], [2773.14569785, 1406.5405226 ], [2806.62667125, 1256.44042836], [3392.99303531, 1137.69520116], [2908.10263735, 1298.38150822], [3615.97183145, 1832.82274977], [4128.46912797, 1593.56296216], [3558.36872853, 2064.51618063], [2885.45082314, 1509.32652806], [3106.67717295, 1903.49557987], [3439.6617577 , 2276.44773767], [2628.22670423, 2051.78724561], [2789.70873377, 1806.36653735], [2665.39954766, 1885.90842814]])\n", "\n", "cam_group.addPointCorrespondenceInformation(corresponding1, corresponding2)" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "We can now run a metropolis sampling to obtain the parameters for the fit. The setup has 5 degrees of freedom (in total 6, but we have fixed one already with the baseline). We keep the tilt of the first camera constant, as this tilt and the direction of the baseline define the orientation of the coordinate system in space." ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "100%|████████████████████████████████████████| 1000000/1000000 [36:55<00:00, 451.28it/s, acc_rate=0.37, factor=0.00363]\n" ] } ], "source": [ "trace = cam_group.metropolis([\n", " ct.FitParameter(\"C0_heading_deg\", lower=0, upper=45, value=15),\n", " ct.FitParameter(\"C0_roll_deg\", lower=-5, upper=5, value=0),\n", " ct.FitParameter(\"C1_tilt_deg\", lower=45, upper=135, value=85),\n", " ct.FitParameter(\"C1_heading_deg\", lower=-45, upper=0, value=-15),\n", " ct.FitParameter(\"C1_roll_deg\", lower=-5, upper=5, value=0)\n", " ], iterations=1e6)" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "The result can be saved to a csv file:" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [], "source": [ "trace[::100].to_csv(\"trace_stereo.csv\", index=False)" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "And loaded again:" ] }, { "cell_type": "code", "execution_count": 8, "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "trace = pd.read_csv(\"trace_stereo.csv\")\n", "cam_group.set_trace(trace)" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "And the traces can be ploted:" ] }, { "cell_type": "code", "execution_count": 9, "metadata": {}, "outputs": [ { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "cam_group.plotTrace()" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "## Measuring" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "This section will use point correspondences in a stereo rig to measure distances. We define the start points and the end points of the distances we want to measure in both images:" ] }, { "cell_type": "code", "execution_count": 10, "metadata": {}, "outputs": [], "source": [ "points1_start = np.array([[4186.75025841, 1275.33146571], [2331.92641872, 1019.27786453], [2001.20914719, 644.51861935], [1133.13081198, 1109.00803712], [ 900.20742144, 1016.83613865], [3287.47982902, 882.47189246], [3965.52607095, 693.78282123], [2689.04269923, 1033.94065467], [2889.82546079, 1144.44899631]])\n", "points1_end = np.array([[3833.11271185, 1441.03505624], [2234.96105241, 1095.94815417], [2141.25664041, 614.81157533], [1797.43541128, 705.61811479], [1594.1321895 , 630.02825115], [4107.48373669, 942.87002974], [3238.54411195, 640.438481 ], [2868.55390207, 951.44851233], [3070.37430064, 1053.65575788]])\n", "\n", "points2_start = np.array([[3390.15132692, 2262.6779465 ], [2312.57037143, 1067.65991742], [2112.03817032, 583.67614881], [ 910.64487701, 538.92782709], [ 955.7855831 , 454.96068822], [2907.87901391, 1297.11142789], [4125.81321176, 1596.76190514], [2524.07911816, 1219.84947435], [2472.59638039, 1365.81424465]])\n", "points2_end = np.array([[2674.06029559, 2066.93032418], [2136.89506898, 1071.98893465], [2200.59552941, 598.28354825], [1706.9610645 , 517.94260225], [1718.87393798, 432.45627336], [3612.29928906, 1832.37417555], [3393.60277139, 1137.62004484], [2807.91089842, 1255.11024902], [2773.34835691, 1405.2303788 ]])\n" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "This allows to calculate the 3D points of the start and end points and measure the distance between them:" ] }, { "cell_type": "code", "execution_count": 11, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "[0.13885105 0.04962405 0.03691601 0.21049065 0.21220163 0.21331054\n", " 0.21138547 0.07524718 0.0751257 ]\n" ] } ], "source": [ "points_start_3D = cam_group.spaceFromImages(points1_start, points2_start)\n", "points_end_3D = cam_group.spaceFromImages(points1_end, points2_end)\n", "\n", "distances = np.linalg.norm(points_end_3D-points_start_3D, axis=-1)\n", "print(distances)" ] }, { "cell_type": "raw", "metadata": { "raw_mimetype": "text/restructuredtext" }, "source": [ "If we do not know the baseline of the camera rig, we can now use one of those distances to scale the setup:" ] }, { "cell_type": "code", "execution_count": 12, "metadata": {}, "outputs": [], "source": [ "# the first distance is known to be 14cm\n", "scale = 0.14/distances[0]\n", "# we use this ratio to scale the space\n", "cam_group.scaleSpace(scale)" ] } ], "metadata": { "celltoolbar": "Raw Cell Format", "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": "3.7.1" } }, "nbformat": 4, "nbformat_minor": 1 }