KITTI_Pseudo_Lidar.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "KITTI_Pseudo_Lidar",
      "provenance": [],
      "collapsed_sections": [
        "B55FDtx-SR8J"
      ],
      "toc_visible": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "k2VGHs6bOOF0",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 139
        },
        "outputId": "5b43162e-ee1d-4b6a-c396-5246aff1618f"
      },
      "source": [
        "!git clone https://github.com/mileyan/pseudo_lidar_V2"
      ],
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Cloning into 'pseudo_lidar_V2'...\n",
            "remote: Enumerating objects: 13, done.\u001b[K\n",
            "remote: Counting objects: 100% (13/13), done.\u001b[K\n",
            "remote: Compressing objects: 100% (13/13), done.\u001b[K\n",
            "remote: Total 100 (delta 4), reused 0 (delta 0), pack-reused 87\u001b[K\n",
            "Receiving objects: 100% (100/100), 2.89 MiB | 4.80 MiB/s, done.\n",
            "Resolving deltas: 100% (22/22), done.\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-HxLHh7oBiR4",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!mkdir -p pseudo_lidar_V2/kitti/{training,testing}/{calib,image_2,image_3}"
      ],
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "QZmsoh-uCJ4D",
        "colab_type": "text"
      },
      "source": [
        "Upload files from local: calib, left and right images."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "BIL4z-vCCg4t",
        "colab_type": "text"
      },
      "source": [
        "Upload subval2, train2, trainval2 and config file"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RpVJqtDkC935",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "8ddf1a8b-7ef6-4e76-ccd6-889168eaecdb"
      },
      "source": [
        "!mv 007481.txt pseudo_lidar_V2/kitti/testing/calib/ && mv left.png pseudo_lidar_V2/kitti/testing/image_2/007481.png && mv right.png pseudo_lidar_V2/kitti/testing/image_3/007481.png && mv sdn_kitti_train.config pseudo_lidar_V2/src/configs && mv *.txt pseudo_lidar_V2/split/"
      ],
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "mv: cannot stat 'sdn_kitti_train.config': No such file or directory\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Pk1_2EgDEP4E",
        "colab_type": "text"
      },
      "source": [
        "Upload sdn_kitti_object_trainval.pth"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "30LzHQU8SGyb",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!mkdir -p pseudo_lidar_V2/results/sdn_kitti_train_set/ && mv sdn_kitti_object_trainval.pth pseudo_lidar_V2/results/sdn_kitti_train_set/"
      ],
      "execution_count": 4,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "u6WSRvZROjUU",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "outputId": "ebb0412a-2681-4fa0-b90a-fbf7b4347363"
      },
      "source": [
        "!cd pseudo_lidar_V2/ && ls -la && pip install -r requirements.txt && python ./src/main.py --config ./src/configs/sdn_kitti_test.config --resume ./results/sdn_kitti_train_set/sdn_kitti_object_trainval.pth --dataset kitti --data_list ./split/test2.txt --generate_depth_map --data_tag test"
      ],
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "total 64\n",
            "drwxr-xr-x 9 root root  4096 Jun 24 09:13 .\n",
            "drwxr-xr-x 1 root root  4096 Jun 24 10:27 ..\n",
            "drwxr-xr-x 2 root root  4096 Jun 24 08:58 figures\n",
            "drwxr-xr-x 5 root root  4096 Jun 24 08:58 gdc\n",
            "drwxr-xr-x 8 root root  4096 Jun 24 08:58 .git\n",
            "-rw-r--r-- 1 root root  1217 Jun 24 08:58 .gitignore\n",
            "drwxr-xr-x 4 root root  4096 Jun 24 08:58 kitti\n",
            "-rw-r--r-- 1 root root  1065 Jun 24 08:58 LICENSE\n",
            "-rw-r--r-- 1 root root 11092 Jun 24 08:58 README.md\n",
            "-rw-r--r-- 1 root root    71 Jun 24 08:58 requirements.txt\n",
            "drwxr-xr-x 3 root root  4096 Jun 24 09:13 results\n",
            "-rw-r--r-- 1 root root  1369 Jun 24 08:58 sceneflow.py\n",
            "drwxr-xr-x 2 root root  4096 Jun 24 09:29 split\n",
            "drwxr-xr-x 8 root root  4096 Jun 24 09:13 src\n",
            "Requirement already satisfied: configargparse==1. in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 1)) (1.0)\n",
            "Requirement already satisfied: opencv-python in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 2)) (4.1.2.30)\n",
            "Requirement already satisfied: scipy in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 3)) (1.4.1)\n",
            "Requirement already satisfied: tensorboardx in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 4)) (2.0)\n",
            "Requirement already satisfied: tqdm in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 5)) (4.41.1)\n",
            "Requirement already satisfied: pillow==6.2.0 in /usr/local/lib/python3.6/dist-packages (from -r requirements.txt (line 6)) (6.2.0)\n",
            "Requirement already satisfied: numpy>=1.11.3 in /usr/local/lib/python3.6/dist-packages (from opencv-python->-r requirements.txt (line 2)) (1.18.5)\n",
            "Requirement already satisfied: protobuf>=3.8.0 in /usr/local/lib/python3.6/dist-packages (from tensorboardx->-r requirements.txt (line 4)) (3.10.0)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.6/dist-packages (from tensorboardx->-r requirements.txt (line 4)) (1.12.0)\n",
            "Requirement already satisfied: setuptools in /usr/local/lib/python3.6/dist-packages (from protobuf>=3.8.0->tensorboardx->-r requirements.txt (line 4)) (47.3.1)\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     arch: SDNet\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     btrain: 12\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     bval: 4\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     calib_value: 1017\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     checkpoint_interval: 5\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     config: ./src/configs/sdn_kitti_test.config\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     data_list: ./split/test2.txt\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     data_tag: test\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     data_type: depth\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     datapath: ./kitti/testing/\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     dataset: kitti\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     down: 2\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     dynamic_bs: False\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     epochs: 300\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     eval_interval: 50\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     evaluate: False\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     generate_depth_map: True\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     kitti2015: False\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     lr: 0.001\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     lr_gamma: 0.1\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     lr_stepsize: [200]\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     maxdepth: 80\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     maxdisp: 192\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     pretrain: ./results/sdn_kitti_train_set/sdn_kitti_object_trainval.pth\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     resume: ./results/sdn_kitti_train_set/sdn_kitti_object_trainval.pth\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     save_path: ./results/sdn_kitti_train_set\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     split_train: ./split/train2.txt\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     split_val: ./split/subval2.txt\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     start_epoch: 0\n",
            "[2020-06-24 10:27:37 main.py:103] INFO     workers: 8\n",
            "[2020-06-24 10:27:38 main.py:143] INFO     Number of model parameters: 5224768\n",
            "[2020-06-24 10:27:40 main.py:153] INFO     => loading pretrain './results/sdn_kitti_train_set/sdn_kitti_object_trainval.pth'\n",
            "[2020-06-24 10:27:40 main.py:161] INFO     => loading checkpoint './results/sdn_kitti_train_set/sdn_kitti_object_trainval.pth'\n",
            "[2020-06-24 10:27:40 main.py:169] INFO     => loaded checkpoint './results/sdn_kitti_train_set/sdn_kitti_object_trainval.pth' (epoch 300)\n",
            "  0% 0/1 [00:00<?, ?it/s]/usr/local/lib/python3.6/dist-packages/torch/nn/functional.py:2796: UserWarning: nn.functional.upsample is deprecated. Use nn.functional.interpolate instead.\n",
            "  warnings.warn(\"nn.functional.upsample is deprecated. Use nn.functional.interpolate instead.\")\n",
            "/usr/local/lib/python3.6/dist-packages/torch/nn/functional.py:2973: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details.\n",
            "  \"See the documentation of nn.Upsample for details.\".format(mode))\n",
            "/usr/local/lib/python3.6/dist-packages/torch/nn/functional.py:2973: UserWarning: Default upsampling behavior when mode=trilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details.\n",
            "  \"See the documentation of nn.Upsample for details.\".format(mode))\n",
            "100% 1/1 [00:04<00:00,  4.28s/it]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "B55FDtx-SR8J",
        "colab_type": "text"
      },
      "source": [
        "# Datasets"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FCKsApIqUCVm",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 208
        },
        "outputId": "a2a01da5-c4a6-444d-aa39-86aa76c54951"
      },
      "source": [
        "!wget https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_image_2.zip"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2020-06-22 09:29:38--  https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_image_2.zip\n",
            "Resolving s3.eu-central-1.amazonaws.com (s3.eu-central-1.amazonaws.com)... 52.219.73.195\n",
            "Connecting to s3.eu-central-1.amazonaws.com (s3.eu-central-1.amazonaws.com)|52.219.73.195|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 12569945557 (12G) [application/zip]\n",
            "Saving to: ‘data_object_image_2.zip’\n",
            "\n",
            "data_object_image_2 100%[===================>]  11.71G  33.1MB/s    in 4m 45s  \n",
            "\n",
            "2020-06-22 09:34:23 (42.1 MB/s) - ‘data_object_image_2.zip’ saved [12569945557/12569945557]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RSx5hN4RZjgA",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!unzip data_object_image_2.zip"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YCuKzoSnYbEm",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 208
        },
        "outputId": "283d25ab-e57e-4eca-ae6c-5da9c181dc79"
      },
      "source": [
        "!wget https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_image_3.zip"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "--2020-06-22 09:55:41--  https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_image_3.zip\n",
            "Resolving s3.eu-central-1.amazonaws.com (s3.eu-central-1.amazonaws.com)... 52.219.72.107\n",
            "Connecting to s3.eu-central-1.amazonaws.com (s3.eu-central-1.amazonaws.com)|52.219.72.107|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 12002946646 (11G) [application/zip]\n",
            "Saving to: ‘data_object_image_3.zip’\n",
            "\n",
            "data_object_image_3 100%[===================>]  11.18G  36.3MB/s    in 4m 30s  \n",
            "\n",
            "2020-06-22 10:00:12 (42.3 MB/s) - ‘data_object_image_3.zip’ saved [12002946646/12002946646]\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "RSZD3X0ondYB",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!rm pseudo_lidar_V2/results/sdn_kitti_train_set/pseudo_lidar_trainval/*"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Rt_-29T9dLUv",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "!unzip data_object_image_3.zip"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1_a3FYoJSZCO",
        "colab_type": "text"
      },
      "source": [
        "# Postprocess & Analysis"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "R6tZ7UHMoDqS",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 153
        },
        "outputId": "3ed2f401-82ea-4453-d6de-1b0a61326a91"
      },
      "source": [
        "import numpy as np\n",
        "from matplotlib import pyplot as plt\n",
        "\n",
        "img_array = np.load('pseudo_lidar_V2/results/sdn_kitti_train_set/depth_maps/test/007481.npy')\n",
        "\n",
        "plt.imshow(img_array, cmap='gray')\n",
        "plt.show()"
      ],
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-zVMtaiSo50s",
        "colab_type": "code",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        },
        "outputId": "7e7f37a5-00fc-413b-c93b-9ce8986d17a3"
      },
      "source": [
        "!cd pseudo_lidar_V2/ && python ./src/preprocess/generate_lidar_from_depth.py --calib_dir ./kitti/testing/calib/ --depth_dir ./results/sdn_kitti_train_set/depth_maps/test/ --save_dir ./results/sdn_kitti_train_set/pseudo_lidar_trainval/"
      ],
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Finish Depth 007481\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "FVr-OgxbS1uP",
        "colab_type": "text"
      },
      "source": [
        "Data will be available as bin file. Next step converts to .ply file "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7HmjsD-mS4z1",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from pathlib import Path\n",
        "import numpy as np\n",
        "import os.path\n",
        "import math\n",
        "\n",
        "def loadKittiVelodyneFile(file_path, include_luminance = False):\n",
        "    '''\n",
        "    Loads a kitti velodyne file (ex: 000000.bin) into a list of tuples, where each tuple has (x, y, z) or (x, y, z, l)\n",
        "    Right now it discards the 4th vaule of each point, i.e. the luminance\n",
        "    Argument:\n",
        "        - include_luminance: if the function should also store the pont intensisty value in the list of points\n",
        "    '''\n",
        "    # Source: https://github.com/hunse/kitti/blob/master/kitti/velodyne.py\n",
        "    points = np.fromfile(file_path, dtype=np.float32).reshape(-1, 4)\n",
        "    points = points[:, :3]  # exclude luminance\n",
        "\n",
        "    point_tuple_list = []\n",
        "    for i in range(len(points)):\n",
        "        point_tuple_list.append((points[i][0], points[i][1], points[i][2],))\n",
        "\n",
        "    return point_tuple_list\n",
        "\n",
        "def savePlyFile(filepath, tuple_list, attributes = None, color_for_every_point = (0, 255, 0)):\n",
        "    '''\n",
        "    For testing in the Main.py file\n",
        "    Save list of points (possibly with attributes such as color) into a .PLY formated file\n",
        "    Arguments: \n",
        "        - tuple_list: list of points and their attributes\n",
        "        - attributes: to indicate what type of attributes are included in the points:\n",
        "            - c: each point has position + color (r, g, b)\n",
        "    '''\n",
        "    with open(filepath, \"w\") as the_file:\n",
        "        header_lines = [\"ply\", \"format ascii 1.0\"]\n",
        "        header_lines.append(\"element vertex \" + str(len(tuple_list)))\n",
        "        header_lines.append(\"property float x\")\n",
        "        header_lines.append(\"property float y\")\n",
        "        header_lines.append(\"property float z\")\n",
        "\n",
        "        # if point have color \n",
        "        if attributes == \"c\":\n",
        "            header_lines.append(\"property uchar red\")\n",
        "            header_lines.append(\"property uchar green\")\n",
        "            header_lines.append(\"property uchar blue\")\n",
        "        \n",
        "        header_lines.append(\"end_header\")\n",
        "        \n",
        "        for i in range(0, len(header_lines)):\n",
        "            the_file.write(header_lines[i] + \"\\n\")\n",
        "\n",
        "        for i in range(0, len(tuple_list)):\n",
        "            if attributes == \"c\" and len(tuple_list[i]) <= 3:    # if the points dont have color, but the attributes is set to \"c\"\n",
        "                new_tuple = (tuple_list[i][0], tuple_list[i][1], tuple_list[i][2], color_for_every_point[0], color_for_every_point[1], color_for_every_point[2])\n",
        "                the_file.write(tupleToStr(new_tuple) + \"\\n\")\n",
        "            else:\n",
        "                the_file.write(tupleToStr(tuple_list[i]) + \"\\n\")\n",
        "\n",
        "def tupleToStr(tuple):\n",
        "    '''\n",
        "    Converts a tuple of N size into a string, where each element is separated by a space.\n",
        "    Arguments:\n",
        "    - tuple: tuple to be converted into string\n",
        "    Returns:\n",
        "        - string with the tuple values\n",
        "    '''\n",
        "    tuple_string = \"\"\n",
        "    for i in range(0, len(tuple)):\n",
        "        if i == (len(tuple)-1):\n",
        "            tuple_string += str(tuple[i])\n",
        "        else:\n",
        "            tuple_string += str(tuple[i]) + \" \"\n",
        " \n",
        "    return tuple_string"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bmCpZwIgTDOM",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "ptl = loadKittiVelodyneFile('/content/pseudo_lidar_V2/results/sdn_kitti_train_set/pseudo_lidar_trainval/007481.bin')\n",
        "savePlyFile('/content/007481.ply', ptl)"
      ],
      "execution_count": 40,
      "outputs": []
    }
  ]
}