SafeDriveVision_google_colab_version.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "code",
      "metadata": {
        "id": "Fj9YcAnsT4B_"
      },
      "source": [
        "\n",
        "from IPython.display import display, Javascript, Image\n",
        "from google.colab.output import eval_js\n",
        "from base64 import b64decode, b64encode\n",
        "import cv2\n",
        "import numpy as np\n",
        "import PIL\n",
        "import io\n",
        "\n",
        "import html\n",
        "import time\n",
        "import dlib"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# prompt: print la version de dlib\n",
        "\n",
        "print(dlib.__version__)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Qmf7BrKBQdtR",
        "outputId": "8002ded3-6c81-4b20-a198-acaba7277875"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "19.24.4\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!git clone https://github.com/ultralytics/yolov5  # clone\n",
        "%cd yolov5\n",
        "%pip install -qr requirements.txt comet_ml  # install\n",
        "\n",
        "import torch\n",
        "\n",
        "import utils\n",
        "display = utils.notebook_init()  # checks"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "kB8UhQuGD4nu",
        "outputId": "9707140d-e9a1-47b9-c585-45414e014f15"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "YOLOv5 🚀 v7.0-307-g920c721e Python-3.10.12 torch-2.2.1+cu121 CUDA:0 (Tesla T4, 15102MiB)\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Setup complete ✅ (2 CPUs, 12.7 GB RAM, 29.9/78.2 GB disk)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "\n",
        "# Charger un modèle pré-entraîné depuis Ultralytics\n",
        "model = torch.hub.load('ultralytics/yolov5', 'yolov5s', pretrained=True)  # Ici, 'yolov5s' peut être remplacé par 'yolov5m', 'yolov5l', ou 'yolov5x' selon le modèle désiré\n",
        "# Afficher les noms des classes\n",
        "print(model.names)\n",
        "# Supposons que model.names soit un dictionnaire\n",
        "# Vous devez d'abord comprendre comment les données sont structurées dans ce dictionnaire.\n",
        "print(model.names)  # Imprimez pour comprendre la structure\n",
        "\n",
        "# Supposons que model.names soit un dictionnaire\n",
        "# Vous devez d'abord comprendre comment les données sont structurées dans ce dictionnaire.\n",
        "print(model.names)  # Imprimez pour comprendre la structure\n",
        "\n",
        "# Si model.names est un dictionnaire et contient les noms des classes sous une certaine clé\n",
        "if isinstance(model.names, dict):\n",
        "    class_names = list(model.names.values())  # ou list(model.names.keys()) selon la structure\n",
        "    if 'cell phone' in class_names:\n",
        "        cell_phone_index = class_names.index('cell phone')\n",
        "        print(\"L'index pour la classe 'cell phone' est :\", cell_phone_index)\n",
        "    else:\n",
        "        print(\"La classe 'cell phone' n'est pas dans les classes du modèle.\")\n",
        "else:\n",
        "    # Si c'est une liste, utilisez directement\n",
        "    if 'cell phone' in model.names:\n",
        "        cell_phone_index = model.names.index('cell phone')\n",
        "        print(\"L'index pour la classe 'cell phone' est :\", cell_phone_index)\n",
        "    else:\n",
        "        print(\"La classe 'cell phone' n'est pas dans les classes du modèle.\")\n",
        "\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KZ-oEcuncFp4",
        "outputId": "2da59378-91a2-45e3-b5a6-bd021b664231"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Using cache found in /root/.cache/torch/hub/ultralytics_yolov5_master\n",
            "YOLOv5 🚀 v7.0-307-g920c721e Python-3.10.12 torch-2.2.1+cu121 CUDA:0 (Tesla T4, 15102MiB)\n",
            "\n",
            "Fusing layers... \n",
            "YOLOv5s summary: 213 layers, 7225885 parameters, 0 gradients, 16.4 GFLOPs\n",
            "Adding AutoShape... \n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "{0: 'person', 1: 'bicycle', 2: 'car', 3: 'motorcycle', 4: 'airplane', 5: 'bus', 6: 'train', 7: 'truck', 8: 'boat', 9: 'traffic light', 10: 'fire hydrant', 11: 'stop sign', 12: 'parking meter', 13: 'bench', 14: 'bird', 15: 'cat', 16: 'dog', 17: 'horse', 18: 'sheep', 19: 'cow', 20: 'elephant', 21: 'bear', 22: 'zebra', 23: 'giraffe', 24: 'backpack', 25: 'umbrella', 26: 'handbag', 27: 'tie', 28: 'suitcase', 29: 'frisbee', 30: 'skis', 31: 'snowboard', 32: 'sports ball', 33: 'kite', 34: 'baseball bat', 35: 'baseball glove', 36: 'skateboard', 37: 'surfboard', 38: 'tennis racket', 39: 'bottle', 40: 'wine glass', 41: 'cup', 42: 'fork', 43: 'knife', 44: 'spoon', 45: 'bowl', 46: 'banana', 47: 'apple', 48: 'sandwich', 49: 'orange', 50: 'broccoli', 51: 'carrot', 52: 'hot dog', 53: 'pizza', 54: 'donut', 55: 'cake', 56: 'chair', 57: 'couch', 58: 'potted plant', 59: 'bed', 60: 'dining table', 61: 'toilet', 62: 'tv', 63: 'laptop', 64: 'mouse', 65: 'remote', 66: 'keyboard', 67: 'cell phone', 68: 'microwave', 69: 'oven', 70: 'toaster', 71: 'sink', 72: 'refrigerator', 73: 'book', 74: 'clock', 75: 'vase', 76: 'scissors', 77: 'teddy bear', 78: 'hair drier', 79: 'toothbrush'}\n",
            "{0: 'person', 1: 'bicycle', 2: 'car', 3: 'motorcycle', 4: 'airplane', 5: 'bus', 6: 'train', 7: 'truck', 8: 'boat', 9: 'traffic light', 10: 'fire hydrant', 11: 'stop sign', 12: 'parking meter', 13: 'bench', 14: 'bird', 15: 'cat', 16: 'dog', 17: 'horse', 18: 'sheep', 19: 'cow', 20: 'elephant', 21: 'bear', 22: 'zebra', 23: 'giraffe', 24: 'backpack', 25: 'umbrella', 26: 'handbag', 27: 'tie', 28: 'suitcase', 29: 'frisbee', 30: 'skis', 31: 'snowboard', 32: 'sports ball', 33: 'kite', 34: 'baseball bat', 35: 'baseball glove', 36: 'skateboard', 37: 'surfboard', 38: 'tennis racket', 39: 'bottle', 40: 'wine glass', 41: 'cup', 42: 'fork', 43: 'knife', 44: 'spoon', 45: 'bowl', 46: 'banana', 47: 'apple', 48: 'sandwich', 49: 'orange', 50: 'broccoli', 51: 'carrot', 52: 'hot dog', 53: 'pizza', 54: 'donut', 55: 'cake', 56: 'chair', 57: 'couch', 58: 'potted plant', 59: 'bed', 60: 'dining table', 61: 'toilet', 62: 'tv', 63: 'laptop', 64: 'mouse', 65: 'remote', 66: 'keyboard', 67: 'cell phone', 68: 'microwave', 69: 'oven', 70: 'toaster', 71: 'sink', 72: 'refrigerator', 73: 'book', 74: 'clock', 75: 'vase', 76: 'scissors', 77: 'teddy bear', 78: 'hair drier', 79: 'toothbrush'}\n",
            "{0: 'person', 1: 'bicycle', 2: 'car', 3: 'motorcycle', 4: 'airplane', 5: 'bus', 6: 'train', 7: 'truck', 8: 'boat', 9: 'traffic light', 10: 'fire hydrant', 11: 'stop sign', 12: 'parking meter', 13: 'bench', 14: 'bird', 15: 'cat', 16: 'dog', 17: 'horse', 18: 'sheep', 19: 'cow', 20: 'elephant', 21: 'bear', 22: 'zebra', 23: 'giraffe', 24: 'backpack', 25: 'umbrella', 26: 'handbag', 27: 'tie', 28: 'suitcase', 29: 'frisbee', 30: 'skis', 31: 'snowboard', 32: 'sports ball', 33: 'kite', 34: 'baseball bat', 35: 'baseball glove', 36: 'skateboard', 37: 'surfboard', 38: 'tennis racket', 39: 'bottle', 40: 'wine glass', 41: 'cup', 42: 'fork', 43: 'knife', 44: 'spoon', 45: 'bowl', 46: 'banana', 47: 'apple', 48: 'sandwich', 49: 'orange', 50: 'broccoli', 51: 'carrot', 52: 'hot dog', 53: 'pizza', 54: 'donut', 55: 'cake', 56: 'chair', 57: 'couch', 58: 'potted plant', 59: 'bed', 60: 'dining table', 61: 'toilet', 62: 'tv', 63: 'laptop', 64: 'mouse', 65: 'remote', 66: 'keyboard', 67: 'cell phone', 68: 'microwave', 69: 'oven', 70: 'toaster', 71: 'sink', 72: 'refrigerator', 73: 'book', 74: 'clock', 75: 'vase', 76: 'scissors', 77: 'teddy bear', 78: 'hair drier', 79: 'toothbrush'}\n",
            "L'index pour la classe 'cell phone' est : 67\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "09b_0FAnUa9y"
      },
      "source": [
        "\n",
        "def js_to_image(js_reply):\n",
        "  \"\"\"\n",
        "  Params:\n",
        "          js_reply: JavaScript object containing image from webcam\n",
        "  Returns:\n",
        "          img: OpenCV BGR image\n",
        "  \"\"\"\n",
        "\n",
        "  image_bytes = b64decode(js_reply.split(',')[1])\n",
        "\n",
        "  jpg_as_np = np.frombuffer(image_bytes, dtype=np.uint8)\n",
        "\n",
        "  img = cv2.imdecode(jpg_as_np, flags=1)\n",
        "\n",
        "  return img\n",
        "\n",
        "# function to convert OpenCV Rectangle bounding box image into base64 byte string to be overlayed on video stream\n",
        "def bbox_to_bytes(bbox_array):\n",
        "  \"\"\"\n",
        "  Params:\n",
        "          bbox_array: Numpy array (pixels) containing rectangle to overlay on video stream.\n",
        "  Returns:\n",
        "        bytes: Base64 image byte string\n",
        "  \"\"\"\n",
        "\n",
        "  bbox_PIL = PIL.Image.fromarray(bbox_array, 'RGBA')\n",
        "  iobuf = io.BytesIO()\n",
        "\n",
        "  bbox_PIL.save(iobuf, format='png')\n",
        "\n",
        "  bbox_bytes = 'data:image/png;base64,{}'.format((str(b64encode(iobuf.getvalue()), 'utf-8')))\n",
        "\n",
        "  return bbox_bytes"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ZpA68lTrcvZs"
      },
      "source": [
        "face_cascade = cv2.CascadeClassifier(cv2.samples.findFile(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml'))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "XwUtEojjnPRb"
      },
      "source": [
        "def take_photo(filename='photo.jpg', quality=0.8):\n",
        "  js = Javascript('''\n",
        "    async function takePhoto(quality) {\n",
        "      const div = document.createElement('div');\n",
        "      const capture = document.createElement('button');\n",
        "      capture.textContent = 'Capture';\n",
        "      div.appendChild(capture);\n",
        "\n",
        "      const video = document.createElement('video');\n",
        "      video.style.display = 'block';\n",
        "      const stream = await navigator.mediaDevices.getUserMedia({video: true});\n",
        "\n",
        "      document.body.appendChild(div);\n",
        "      div.appendChild(video);\n",
        "      video.srcObject = stream;\n",
        "      await video.play();\n",
        "\n",
        "      google.colab.output.setIframeHeight(document.documentElement.scrollHeight, true);\n",
        "\n",
        "      await new Promise((resolve) => capture.onclick = resolve);\n",
        "\n",
        "      const canvas = document.createElement('canvas');\n",
        "      canvas.width = video.videoWidth;\n",
        "      canvas.height = video.videoHeight;\n",
        "      canvas.getContext('2d').drawImage(video, 0, 0);\n",
        "      stream.getVideoTracks()[0].stop();\n",
        "      div.remove();\n",
        "      return canvas.toDataURL('image/jpeg', quality);\n",
        "    }\n",
        "    ''')\n",
        "  display(js)\n",
        "\n",
        "\n",
        "  data = eval_js('takePhoto({})'.format(quality))\n",
        "\n",
        "  img = js_to_image(data)\n",
        "\n",
        "  gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)\n",
        "  print(gray.shape)\n",
        "\n",
        "  faces = face_cascade.detectMultiScale(gray)\n",
        "\n",
        "  for (x,y,w,h) in faces:\n",
        "      img = cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)\n",
        "\n",
        "  cv2.imwrite(filename, img)\n",
        "\n",
        "  return filename"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ilLkpcKanPRb",
        "outputId": "08b44696-3920-4b8d-d709-6380cf0606fc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 532
        }
      },
      "source": [
        "try:\n",
        "  filename = take_photo('photo.jpg')\n",
        "  print('Saved to {}'.format(filename))\n",
        "  display(Image(filename))\n",
        "except Exception as err:\n",
        "\n",
        "  print(str(err))"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "    async function takePhoto(quality) {\n",
              "      const div = document.createElement('div');\n",
              "      const capture = document.createElement('button');\n",
              "      capture.textContent = 'Capture';\n",
              "      div.appendChild(capture);\n",
              "\n",
              "      const video = document.createElement('video');\n",
              "      video.style.display = 'block';\n",
              "      const stream = await navigator.mediaDevices.getUserMedia({video: true});\n",
              "\n",
              "      document.body.appendChild(div);\n",
              "      div.appendChild(video);\n",
              "      video.srcObject = stream;\n",
              "      await video.play();\n",
              "\n",
              "      google.colab.output.setIframeHeight(document.documentElement.scrollHeight, true);\n",
              "\n",
              "      await new Promise((resolve) => capture.onclick = resolve);\n",
              "\n",
              "      const canvas = document.createElement('canvas');\n",
              "      canvas.width = video.videoWidth;\n",
              "      canvas.height = video.videoHeight;\n",
              "      canvas.getContext('2d').drawImage(video, 0, 0);\n",
              "      stream.getVideoTracks()[0].stop();\n",
              "      div.remove();\n",
              "      return canvas.toDataURL('image/jpeg', quality);\n",
              "    }\n",
              "    "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "(480, 640)\n",
            "Saved to photo.jpg\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "image/jpeg": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import math\n",
        "import cv2\n",
        "\n",
        "# 3D model points.\n",
        "model_points = np.array([\n",
        "    (0.0, 0.0, 0.0),\n",
        "    (0.0, -330.0, -65.0),\n",
        "    (-225.0, 170.0, -135.0),\n",
        "    (225.0, 170.0, -135.0),\n",
        "    (-150.0, -150.0, -125.0),\n",
        "    (150.0, -150.0, -125.0)\n",
        "])\n",
        "\n",
        "def isRotationMatrix(R):\n",
        "    Rt = np.transpose(R)\n",
        "    shouldBeIdentity = np.dot(Rt, R)\n",
        "    I = np.identity(3, dtype=R.dtype)\n",
        "    n = np.linalg.norm(I - shouldBeIdentity)\n",
        "    return n < 1e-6\n",
        "\n",
        "def rotationMatrixToEulerAngles(R):\n",
        "    assert(isRotationMatrix(R))\n",
        "    sy = math.sqrt(R[0, 0] * R[0, 0] + R[1, 0] * R[1, 0])\n",
        "    singular = sy < 1e-6\n",
        "    if not singular:\n",
        "        x = math.atan2(R[2, 1], R[2, 2])\n",
        "        y = math.atan2(-R[2, 0], sy)\n",
        "        z = math.atan2(R[1, 0], R[0, 0])\n",
        "    else:\n",
        "        x = math.atan2(-R[1, 2], R[1, 1])\n",
        "        y = math.atan2(-R[2, 0], sy)\n",
        "        z = 0\n",
        "    return np.array([x, y, z])\n",
        "\n",
        "\n",
        "def getHeadTiltAndCoords(size, image_points, frame_height):\n",
        "    focal_length = size[1]\n",
        "    center = (size[1]/2, size[0]/2)\n",
        "    camera_matrix = np.array([[focal_length, 0, center[0]], [\n",
        "        0, focal_length, center[1]], [0, 0, 1]], dtype=\"double\")\n",
        "\n",
        "    dist_coeffs = np.zeros((4, 1))\n",
        "    (_, rotation_vector, translation_vector) = cv2.solvePnP(model_points, image_points,\n",
        "                                                                  camera_matrix, dist_coeffs,\n",
        "                                                                  flags = cv2.SOLVEPNP_ITERATIVE)\n",
        "    (nose_end_point2D, _) = cv2.projectPoints(np.array(\n",
        "        [(0.0, 0.0, 1000.0)]), rotation_vector, translation_vector, camera_matrix, dist_coeffs)\n",
        "\n",
        "    rotation_matrix, _ = cv2.Rodrigues(rotation_vector)\n",
        "    head_tilt_degree = abs(\n",
        "        [-180] - np.rad2deg([rotationMatrixToEulerAngles(rotation_matrix)[0]]))\n",
        "    starting_point = (int(image_points[0][0]), int(image_points[0][1]))\n",
        "    ending_point = (int(nose_end_point2D[0][0][0]), int(nose_end_point2D[0][0][1]))\n",
        "\n",
        "    ending_point_alternate = (ending_point[0], frame_height // 2)\n",
        "\n",
        "    return head_tilt_degree, starting_point, ending_point, ending_point_alternate\n"
      ],
      "metadata": {
        "id": "nIQqUQe5lFp9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from scipy.spatial import distance as dist\n",
        "\n",
        "def mouth_aspect_ratio(mouth):\n",
        "    A = dist.euclidean(mouth[2], mouth[10])\n",
        "    B = dist.euclidean(mouth[4], mouth[8])\n",
        "    C = dist.euclidean(mouth[0], mouth[6])\n",
        "    mar = (A + B) / (2.0 * C)\n",
        "    return mar"
      ],
      "metadata": {
        "id": "Jlbw7nr8lGhD"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from scipy.spatial import distance as dist\n",
        "\n",
        "def eye_aspect_ratio(eye):\n",
        "    A = dist.euclidean(eye[1], eye[5])\n",
        "    B = dist.euclidean(eye[2], eye[4])\n",
        "    C = dist.euclidean(eye[0], eye[3])\n",
        "    ear = (A + B) / (2.0 * C)\n",
        "    return ear\n",
        "\n"
      ],
      "metadata": {
        "id": "c04QLeTRlJXW"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "ghUlAJzKSjFT"
      },
      "source": [
        "from IPython.display import display, Javascript\n",
        "\n",
        "def video_stream():\n",
        "    js = Javascript('''\n",
        "        var video;\n",
        "        var div = null;\n",
        "        var stream;\n",
        "        var captureCanvas;\n",
        "        var imgElement;\n",
        "        var labelElement;\n",
        "\n",
        "        var pendingResolve = null;\n",
        "        var shutdown = false;\n",
        "\n",
        "        function removeDom() {\n",
        "            stream.getVideoTracks()[0].stop();\n",
        "            video.remove();\n",
        "            div.remove();\n",
        "            video = null;\n",
        "            div = null;\n",
        "            stream = null;\n",
        "            imgElement = null;\n",
        "            captureCanvas = null;\n",
        "            labelElement = null;\n",
        "        }\n",
        "\n",
        "        function onAnimationFrame() {\n",
        "            if (!shutdown) {\n",
        "                window.requestAnimationFrame(onAnimationFrame);\n",
        "            }\n",
        "            if (pendingResolve) {\n",
        "                var result = \"\";\n",
        "                if (!shutdown) {\n",
        "                    captureCanvas.getContext('2d').drawImage(video, 0, 0, 640, 480);\n",
        "                    result = captureCanvas.toDataURL('image/jpeg', 0.8);\n",
        "                }\n",
        "                var lp = pendingResolve;\n",
        "                pendingResolve = null;\n",
        "                lp(result);\n",
        "            }\n",
        "        }\n",
        "\n",
        "        async function createDom() {\n",
        "            if (div !== null) {\n",
        "                return stream;\n",
        "            }\n",
        "\n",
        "            div = document.createElement('div');\n",
        "            div.style.border = '2px solid black';\n",
        "            div.style.padding = '3px';\n",
        "            div.style.width = '100%';\n",
        "            div.style.maxWidth = '600px';\n",
        "            document.body.appendChild(div);\n",
        "\n",
        "            const modelOut = document.createElement('div');\n",
        "            modelOut.innerHTML = \"Status:\";\n",
        "            labelElement = document.createElement('span');\n",
        "            labelElement.innerText = 'No data';\n",
        "            labelElement.style.fontWeight = 'bold';\n",
        "            modelOut.appendChild(labelElement);\n",
        "            div.appendChild(modelOut);\n",
        "\n",
        "            video = document.createElement('video');\n",
        "            video.style.display = 'block';\n",
        "            video.width = div.clientWidth - 6;\n",
        "            video.setAttribute('playsinline', '');\n",
        "            video.onclick = () => { shutdown = true; };\n",
        "            stream = await navigator.mediaDevices.getUserMedia({video: { facingMode: \"environment\"}});\n",
        "            div.appendChild(video);\n",
        "\n",
        "            imgElement = document.createElement('img');\n",
        "            imgElement.style.position = 'absolute';\n",
        "            imgElement.style.zIndex = 1;\n",
        "            imgElement.onclick = () => { shutdown = true; };\n",
        "            div.appendChild(imgElement);\n",
        "\n",
        "            const instruction = document.createElement('div');\n",
        "            instruction.innerHTML = 'When finished, click here or on the video to stop this demo';\n",
        "            div.appendChild(instruction);\n",
        "            instruction.onclick = () => { shutdown = true; };\n",
        "\n",
        "            video.srcObject = stream;\n",
        "            await video.play();\n",
        "\n",
        "            captureCanvas = document.createElement('canvas');\n",
        "            captureCanvas.width = 640;\n",
        "            captureCanvas.height = 480;\n",
        "            window.requestAnimationFrame(onAnimationFrame);\n",
        "\n",
        "            return stream;\n",
        "        }\n",
        "\n",
        "        async function stream_frame(label, imgData) {\n",
        "            if (shutdown) {\n",
        "                removeDom();\n",
        "                shutdown = false;\n",
        "                return '';\n",
        "            }\n",
        "\n",
        "            var preCreate = Date.now();\n",
        "            stream = await createDom();\n",
        "\n",
        "            var preShow = Date.now();\n",
        "            if (label != \"\") {\n",
        "                labelElement.innerHTML = label;\n",
        "            }\n",
        "\n",
        "            if (imgData != \"\") {\n",
        "                var videoRect = video.getClientRects()[0];\n",
        "                imgElement.style.top = videoRect.top + \"px\";\n",
        "                imgElement.style.left = videoRect.left + \"px\";\n",
        "                imgElement.style.width = videoRect.width + \"px\";\n",
        "                imgElement.style.height = videoRect.height + \"px\";\n",
        "                imgElement.src = imgData;\n",
        "            }\n",
        "\n",
        "            var preCapture = Date.now();\n",
        "            var result = await new Promise(function(resolve, reject) {\n",
        "                pendingResolve = resolve;\n",
        "            });\n",
        "            shutdown = false;\n",
        "\n",
        "            return {'create': preShow - preCreate, 'show': preCapture - preShow, 'capture': Date.now() - preCapture, 'img': result};\n",
        "        }\n",
        "    ''')\n",
        "    display(js)\n",
        "\n",
        "def video_frame(label, bbox):\n",
        "    data = eval_js(f'stream_frame(\"{label}\", \"{bbox}\")')\n",
        "    return data\n"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def nose_aspect_ratio(nose):\n",
        "    vertical_distance = dist.euclidean(nose[0], nose[2])\n",
        "    depth_distance = dist.euclidean(nose[0], nose[1])\n",
        "    nar = depth_distance / vertical_distance\n",
        "\n",
        "    return nar"
      ],
      "metadata": {
        "id": "ney_upF5cMd8"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "\n",
        "def calculate_head_angle(eye_left, eye_right, nose_tip):\n",
        "    eye_center = (eye_left + eye_right) / 2\n",
        "    vector_nose = nose_tip - eye_center\n",
        "    vector_horizontal = (eye_right - eye_left)\n",
        "    vector_horizontal[1] = 0\n",
        "    vector_nose_normalized = vector_nose / np.linalg.norm(vector_nose)\n",
        "    vector_horizontal_normalized = vector_horizontal / np.linalg.norm(vector_horizontal)\n",
        "    angle_rad = np.arccos(np.clip(np.dot(vector_nose_normalized, vector_horizontal_normalized), -1.0, 1.0))\n",
        "    angle_deg = np.degrees(angle_rad)\n",
        "\n",
        "    return angle_deg\n"
      ],
      "metadata": {
        "id": "-yXnMFcNkH0r"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "old_gray = None\n",
        "p0 = None\n",
        "lk_params = {\n",
        "    'winSize': (15, 15),\n",
        "    'maxLevel': 2,\n",
        "    'criteria': (cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_COUNT, 10, 0.03)\n",
        "}\n",
        "\n",
        "def track_landmarks(img, gray, landmarks_points):\n",
        "    global old_gray, p0\n",
        "    if old_gray is None:\n",
        "        p0 = np.array([[x, y] for (x, y) in landmarks_points], dtype=np.float32)\n",
        "        old_gray = gray.copy()\n",
        "        return landmarks_points\n",
        "\n",
        "    p1, st, err = cv2.calcOpticalFlowPyrLK(old_gray, gray, p0, None, **lk_params)\n",
        "    old_gray = gray.copy()\n",
        "    if p1 is not None and st.flatten().any():\n",
        "        p1 = p1[st.flatten() == 1]\n",
        "        p0 = p1.reshape(-1, 1, 2)\n",
        "        return p1.reshape(-1, 2).tolist()\n",
        "    else:\n",
        "        p0 = None\n",
        "        return []\n",
        "\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "3zzhOfGghWeb"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from imutils import face_utils\n",
        "import argparse\n",
        "import imutils\n",
        "from IPython.display import HTML, display\n",
        "video_stream()\n",
        "time.sleep(2.0)\n",
        "from IPython.display import Audio, display\n",
        "\n",
        "s1= '/content/ouvrez_les_yeux.mp3'\n",
        "s2 = '/content/regardez_devant.mp3'\n",
        "s3 = '/content/reposer_vous.mp3'\n",
        "def play_sound(audio_file):\n",
        "    audio = Audio(audio_file, autoplay=True)\n",
        "    display(audio)\n",
        "    display(HTML(\"<style>audio{visibility: hidden;}</style>\"))\n",
        "def initialize_trackers(img, landmarks_points):\n",
        "    global multi_tracker\n",
        "    multi_tracker = cv2.legacy.MultiTracker_create()\n",
        "    for (x, y) in landmarks_points:\n",
        "        tracker = cv2.legacy.TrackerCSRT_create()\n",
        "        bbox = (x - 10, y - 10, 20, 20)\n",
        "        multi_tracker.add(tracker, img, bbox)\n",
        "    return True\n",
        "\n",
        "label_html = 'Capturing...'\n",
        "def stop_sound(audio_file):\n",
        "  display(Audio(audio_file, autoplay=False))\n",
        "closed_eye = 0\n",
        "alarm_playing = False\n",
        "bbox = ''\n",
        "COUNTER1 = 0\n",
        "COUNTER2 = 0\n",
        "COUNTER3 = 0\n",
        "EYE_AR_CONSEC_FRAMES = 4\n",
        "(lStart, lEnd) = face_utils.FACIAL_LANDMARKS_IDXS[\"left_eye\"]\n",
        "(rStart, rEnd) = face_utils.FACIAL_LANDMARKS_IDXS[\"right_eye\"]\n",
        "\n",
        "detector = dlib.get_frontal_face_detector()\n",
        "predictor = dlib.shape_predictor('/content/shape_predictor_68_face_landmarks.dat')\n",
        "while True:\n",
        "    js_reply = video_frame(label_html, bbox)\n",
        "    if not js_reply:\n",
        "        break\n",
        "    img = js_to_image(js_reply[\"img\"])\n",
        "    bbox_array = np.zeros([480, 640, 4], dtype=np.uint8)\n",
        "    gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)\n",
        "\n",
        "    faces = detector(gray,0)\n",
        "    eye_closed = False\n",
        "\n",
        "\n",
        "    if len(faces)>0:\n",
        "      text = \"{} face(s) found\".format(len(faces))\n",
        "      cv2.putText(bbox_array, text, (10, 20),\n",
        "                    cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 2)\n",
        "\n",
        "    for face in faces:\n",
        "\n",
        "                font = cv2.FONT_HERSHEY_SIMPLEX\n",
        "                font_scale = 0.7\n",
        "                font_color = (255, 255, 255)\n",
        "                line_type = 2\n",
        "                text_position = (10, 30)\n",
        "                landmarks = predictor(gray, face)\n",
        "                landmarks_points = np.array([(p.x, p.y) for p in landmarks.parts()])\n",
        "                image_points = np.array([\n",
        "                    (landmarks_points[30][0], landmarks_points[30][1]),\n",
        "                    (landmarks_points[8][0], landmarks_points[8][1]),\n",
        "                    (landmarks_points[36][0], landmarks_points[36][1]),\n",
        "                    (landmarks_points[45][0], landmarks_points[45][1]),\n",
        "                    (landmarks_points[48][0], landmarks_points[48][1]),\n",
        "                    (landmarks_points[54][0], landmarks_points[54][1])\n",
        "                ], dtype=\"double\")\n",
        "\n",
        "                (x, y, w, h) = (face.left(), face.top(), face.width(), face.height())\n",
        "                bbox_array = cv2.rectangle(bbox_array,(x,y),(x+w,y+h),(255,0,0),2)\n",
        "\n",
        "                for (x, y) in landmarks_points:\n",
        "                  cv2.circle(bbox_array, (x, y), 2, (255,255,255), -1)\n",
        "                left_eye = landmarks_points[36:42]\n",
        "                right_eye = landmarks_points[42:48]\n",
        "                le_x, le_y, le_w, le_h = cv2.boundingRect(np.array([left_eye]))\n",
        "                re_x, re_y, re_w, re_h = cv2.boundingRect(np.array([right_eye]))\n",
        "                leftEyeHull = cv2.convexHull(np.array([left_eye]))\n",
        "                rightEyeHull = cv2.convexHull(np.array([right_eye]))\n",
        "                cv2.drawContours(bbox_array, [leftEyeHull], -1, (0, 255, 0), 2)\n",
        "                cv2.drawContours(bbox_array, [rightEyeHull], -1, (0, 255, 0), 2)\n",
        "                ear = (eye_aspect_ratio(left_eye) + eye_aspect_ratio(right_eye)) / 2.0\n",
        "                mar = mouth_aspect_ratio(landmarks_points[48:68])\n",
        "                text_y = max(y - 10, 20)\n",
        "                cv2.putText(bbox_array, f'EAR: {ear:.2f}', (10,90), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 255), 2)\n",
        "                cv2.putText(bbox_array, f'MAR: {mar:.2f}', (10,110), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 255), 2)\n",
        "                nose_points = [landmarks_points[27], landmarks_points[30], landmarks_points[33]]\n",
        "                nar = nose_aspect_ratio(nose_points)\n",
        "                text = f'Nose Aspect Ratio: {nar:.2f}'\n",
        "                cv2.putText(bbox_array, text,(10,130),cv2.FONT_HERSHEY_SIMPLEX,0.5,(255,0,0),2)\n",
        "                eye_left = landmarks_points[36]\n",
        "                eye_right = landmarks_points[45]\n",
        "                nose_tip = landmarks_points[33]\n",
        "                head_angle = calculate_head_angle(np.array(eye_left), np.array(eye_right), np.array(nose_tip))\n",
        "                cv2.putText(bbox_array, f'Head Angle: {head_angle:.2f}',(10,70), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 255), 2)\n",
        "                size = img.shape\n",
        "                frame_height = img.shape[0]\n",
        "                head_tilt_degree, start_point, end_point, end_point_alt = getHeadTiltAndCoords(size, image_points, frame_height)\n",
        "                cv2.putText(bbox_array, f'Head Tilt: {head_tilt_degree[0]:.2f} degrees', (10, 50), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 255), 2)\n",
        "                cv2.line(bbox_array, start_point, end_point, (0, 255, 0), 2)\n",
        "                if 0.65 < nar < 0.75:\n",
        "                    cv2.putText(bbox_array, \"good\", (x, y-60), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 2)\n",
        "                    COUNTER2 = 0\n",
        "                else:\n",
        "                    cv2.putText(bbox_array, \"Regardez devant vous !\", (x, y-60), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 2)\n",
        "                    COUNTER2 += 1\n",
        "                    if COUNTER2 >= EYE_AR_CONSEC_FRAMES:\n",
        "                        play_sound(s2)\n",
        "                        COUNTER2 = 0\n",
        "\n",
        "\n",
        "\n",
        "                if ear < 0.2:\n",
        "                            cv2.putText(bbox_array, \"Eyes Closed!\", (x, y - 30), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)\n",
        "                            COUNTER1 += 1\n",
        "                            if COUNTER1 >= EYE_AR_CONSEC_FRAMES:\n",
        "                                play_sound(s1)\n",
        "                                cv2.putText(bbox_array, \"Eyes Closed!\", (x, y - 30), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)\n",
        "                                COUNTER1 = 0\n",
        "                else:\n",
        "                            COUNTER1 = 0\n",
        "\n",
        "                if mar > 0.6:\n",
        "                          play_sound(s3)\n",
        "                          time.sleep(2)\n",
        "                          stop_sound\n",
        "                          cv2.putText(bbox_array, \"Yawning!\", (x, y - 60), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)\n",
        "\n",
        "                head_tilt_degree, start_point, end_point, end_point_alt = getHeadTiltAndCoords(size, image_points, frame_height)\n",
        "\n",
        "                cv2.line(bbox_array, start_point, end_point, (255, 0, 0), 2)\n",
        "                cv2.line(bbox_array, start_point, end_point_alt, (0, 0, 255), 2)\n",
        "\n",
        "    bbox_array[:, :, 3] = (bbox_array.max(axis=2) > 0).astype(int) * 255\n",
        "    bbox_bytes = bbox_to_bytes(bbox_array)\n",
        "    bbox = bbox_bytes\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 76
        },
        "id": "i8lIdSi0Eq_g",
        "outputId": "a2d18589-b770-4ea5-eae9-e225ad1dbfc1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "        var video;\n",
              "        var div = null;\n",
              "        var stream;\n",
              "        var captureCanvas;\n",
              "        var imgElement;\n",
              "        var labelElement;\n",
              "\n",
              "        var pendingResolve = null;\n",
              "        var shutdown = false;\n",
              "\n",
              "        function removeDom() {\n",
              "            stream.getVideoTracks()[0].stop();\n",
              "            video.remove();\n",
              "            div.remove();\n",
              "            video = null;\n",
              "            div = null;\n",
              "            stream = null;\n",
              "            imgElement = null;\n",
              "            captureCanvas = null;\n",
              "            labelElement = null;\n",
              "        }\n",
              "\n",
              "        function onAnimationFrame() {\n",
              "            if (!shutdown) {\n",
              "                window.requestAnimationFrame(onAnimationFrame);\n",
              "            }\n",
              "            if (pendingResolve) {\n",
              "                var result = \"\";\n",
              "                if (!shutdown) {\n",
              "                    captureCanvas.getContext('2d').drawImage(video, 0, 0, 640, 480);\n",
              "                    result = captureCanvas.toDataURL('image/jpeg', 0.8);\n",
              "                }\n",
              "                var lp = pendingResolve;\n",
              "                pendingResolve = null;\n",
              "                lp(result);\n",
              "            }\n",
              "        }\n",
              "\n",
              "        async function createDom() {\n",
              "            if (div !== null) {\n",
              "                return stream;\n",
              "            }\n",
              "\n",
              "            div = document.createElement('div');\n",
              "            div.style.border = '2px solid black';\n",
              "            div.style.padding = '3px';\n",
              "            div.style.width = '100%';\n",
              "            div.style.maxWidth = '600px';\n",
              "            document.body.appendChild(div);\n",
              "\n",
              "            const modelOut = document.createElement('div');\n",
              "            modelOut.innerHTML = \"Status:\";\n",
              "            labelElement = document.createElement('span');\n",
              "            labelElement.innerText = 'No data';\n",
              "            labelElement.style.fontWeight = 'bold';\n",
              "            modelOut.appendChild(labelElement);\n",
              "            div.appendChild(modelOut);\n",
              "\n",
              "            video = document.createElement('video');\n",
              "            video.style.display = 'block';\n",
              "            video.width = div.clientWidth - 6;\n",
              "            video.setAttribute('playsinline', '');\n",
              "            video.onclick = () => { shutdown = true; };\n",
              "            stream = await navigator.mediaDevices.getUserMedia({video: { facingMode: \"environment\"}});\n",
              "            div.appendChild(video);\n",
              "\n",
              "            imgElement = document.createElement('img');\n",
              "            imgElement.style.position = 'absolute';\n",
              "            imgElement.style.zIndex = 1;\n",
              "            imgElement.onclick = () => { shutdown = true; };\n",
              "            div.appendChild(imgElement);\n",
              "\n",
              "            const instruction = document.createElement('div');\n",
              "            instruction.innerHTML = 'When finished, click here or on the video to stop this demo';\n",
              "            div.appendChild(instruction);\n",
              "            instruction.onclick = () => { shutdown = true; };\n",
              "\n",
              "            video.srcObject = stream;\n",
              "            await video.play();\n",
              "\n",
              "            captureCanvas = document.createElement('canvas');\n",
              "            captureCanvas.width = 640;\n",
              "            captureCanvas.height = 480;\n",
              "            window.requestAnimationFrame(onAnimationFrame);\n",
              "\n",
              "            return stream;\n",
              "        }\n",
              "\n",
              "        async function stream_frame(label, imgData) {\n",
              "            if (shutdown) {\n",
              "                removeDom();\n",
              "                shutdown = false;\n",
              "                return '';\n",
              "            }\n",
              "\n",
              "            var preCreate = Date.now();\n",
              "            stream = await createDom();\n",
              "\n",
              "            var preShow = Date.now();\n",
              "            if (label != \"\") {\n",
              "                labelElement.innerHTML = label;\n",
              "            }\n",
              "\n",
              "            if (imgData != \"\") {\n",
              "                var videoRect = video.getClientRects()[0];\n",
              "                imgElement.style.top = videoRect.top + \"px\";\n",
              "                imgElement.style.left = videoRect.left + \"px\";\n",
              "                imgElement.style.width = videoRect.width + \"px\";\n",
              "                imgElement.style.height = videoRect.height + \"px\";\n",
              "                imgElement.src = imgData;\n",
              "            }\n",
              "\n",
              "            var preCapture = Date.now();\n",
              "            var result = await new Promise(function(resolve, reject) {\n",
              "                pendingResolve = resolve;\n",
              "            });\n",
              "            shutdown = false;\n",
              "\n",
              "            return {'create': preShow - preCreate, 'show': preCapture - preShow, 'capture': Date.now() - preCapture, 'img': result};\n",
              "        }\n",
              "    "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.lib.display.Audio object>"
            ],
            "text/html": [
              "\n",
              "                <audio  controls=\"controls\" autoplay=\"autoplay\">\n",
              "                    <source src=\"data:audio/mpeg;base64,SUQzBAAAAAAAI1RTU0UAAAAPAAADTGF2ZjU4LjI5LjEwMAAAAAAAAAAAAAAA//tUwAAAAAAAAAAAAAAAAAAAAAAASW5mbwAAAA8AAAAwAAAkwAAKCg8PFBQaGh8fJCQpKS8vNDQ5OT4+Q0NJSUlOTlNTWFheXmNjaGhtbXJyeHh9fYKCh4eHjY2SkpeXnJyhoaenrKyxsba2vLzBwcbGxsvL0NDW1tvb4ODl5evr8PD19fr6//8AAAAATGF2ZgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAJMDjmNTRAAAAAAAAAAAAAAAAAAAA//tUxKGDUB0i/CelggFToiFE8yARSEIquQHfNhE4lQJJsYzNTXnE0J5tY30KkrvI7s81HHYMHSCQiKlBM9OEDLTJs+zj9zxRIBSpNuhNR2SjpZ3pcavbWotovzi8yGkqFbkzzfFQyKoAaQiAezKFoNRV/3/ooPJhFskETIqJSwhgQgRaGaTJGjNmGxXirdQY3KtwXpGKpTalUYwx2317kFIBkGmKYIFVpLwZixa9FFxekryujqVb4zbFzE02vesS//tUxIwCDAkRESekccl7o6HExJm4fp9z5bB7rloWk6dYziMVCe75agKhcPJZWjg0zFu8TyGEpwTGgu55J7tZqNs5iWISTrKyb5CywpwI4tosSpAjkErCgcHB+ml4ZxHQSJEFwsjSm8hWhpIRjt4j6hmbEwioayBATL0VAEQ36qQAcDKxbtSSIlYLGsdqW49gdD0G5FyvwLjSpgTNJxUUxijWOgnhoGcItmjSoc9hXgeQskoWpVryL7Njxdz9FgWS//tUxIGADMD1EMwNJIldoeNkxI047Z0JKNOLIphQOxINDoea8TaSeYlWPd3GFi54e/cgCinOrBAAgAEceEIqBCJnZBE7rt+IQjdCYEATopD5hDkgBbZII43gA0Z6kT13PJKEIQS6ld7jxCEL4Ss32AzvH19gJgkhpkIYDC1LFj/7+9uQuCkL+XxFRlcu3LLz7n1uft6kQ5IsR5MNYDVaM+k3v//9mVUWPGix3KF7/Vpt/PzvX/8VzcmhRjv5mBoA//tUxHgADD0VFzSUAAozIeanMvABAIGA0AoJ4q2FA8KAQcANhLAShGWpSrTyYInwzcFA6YdcVg2gZVVdCpQdeoEHEzJ1VCy6IXAtNCzuW9rHsIRWsYTIwuhBMC1DkTaRhpSGySYi2JDxIun5MMSu19kpDwy79blTyP3/7/mefhLVEos9HKC7dtyXnIIAgAmAUEEPwecOEv4mumehc+cAp+qZofqjSSZExpSDN6euOBcn1gwuKz6nOk5O5lJAQz4x//tUxFYCDu0dNF20gBopIucdlLIRDsewsBAirFj8S9O3FGagoWDBCOSKVVCcth0lLAvVOauZWRHDAnKqITKqr1c9pzdJ/nClSdjxV5a069AZsa8gG6ktn/RuN2WJVzEOMPJFLyjUag6LSz0QAAgAVWBs0dcmXJ0MAa8zcUBfV0S1wwIkcTtLm+eb5QRmmzdQ+R8oUfzxQwaxU22LE7gVo4ArSHGQGRtkTxxTEPXCEE5nU5LRmZXCBheQVbS5Iw1F//tUxCqAkD0PPwy9jUm4omjphiVwRyupzx0SyPx4pLcXRNt6hUEowM1dY18uOJ5fO101pSC0UFa/DN269K+39DIoCRAAAgrDlA7EgeNx3qikPOzKGHkwRFwJMEZOcSnZuhUdo9KxJnNQarjwzEgIyYiIhKbOJkCkIxXlDJyeZJThX8qYYWMs9gonCZAH7Z1OL0ooJZOJEyTMI9zMlCVPyrhDJyrFpyavfO4Z7672MicJ0pkCSSqKyUg1BstIKDfu//tUxAgACxUbVUwwaVlLmelVpI2wgchYMsmY7IS4lvi9l0+dyvOWcPrgVHB7fcUwMnUqlJl5Sk8xIWHxVn9JZGdQrBW6WZvrC1vVP7boaKT5IhyQ23J3MORMqmwIYMsymlki4NQsRDEKXQEJN6qpBkQQkNmQmK8EoACARwbZUdbFpOUdkf6rL227Kj+1cP85RTvSIZn7//fSxYQUty4Zw8yJFPD5MgmoGn9sDDgMWDSVADABccUaAIFdmR7C/xUg//tUxAeACvUFW6wwS+lWn2sphIohmAOyWHrOb1n+D0PhVC8joBbiWUbV5RlI6oHNlU6SHsclPSwkD8bdmbm0mdtVRbPb+tVOX09dmKdZv1Qo5QzBBAx296HWLOny3AADhDcgYEabCCMSZMYjI5RWmnWCijRXUuMGqPWyqNNEaYDq7x/rxmxoN2ubqveXvXWBVFa9x2Ss6o/NR5afYEQGKDWUSxq2lm//JnVsjzI0kOjhHOxV4CpACABsaYEMvoOt//tUxAYACWT/X6wwSYkjGq109ImuzIRlCZKipElVwwOSbcq4lGOwjE5EBg9f1n4Yi6cOZdWldUYkBI3uM6P1L/+9S7Ek1r/T+e5SHQXOzv6yuPHEYDRRIkJREBg+EPIA5jZK0+BSCaGihODCNaDHlssiwoPxjtpWdhGNU6kV3VxZgpokU+dJ///66s8Nd/+rZEVAIWUFTntJBoS/IkA45GSGoyWBALIM1VG6ymoYzAZyW2fyi71W5gqVUwCPbpMF//tUxBGACVkJYaeMUPEiGih1sYm4Ew7ke1VSMhlb0CiTGR/LK0retflKn/6o6GfUBMZPy2QzlZCtDAS4VAAzrLYW0swDyLHOBYCYDQmsekCoW0UMheNWkkWHGwFRvWUUQqBgIUawVM/4prAacW2lKSrrvX/UDGuys5fN9FZQrrLX0BdV2mogCSxxlRoggPehSD1raqAXG4WUf1fOE/z7xZrtPIZkEIDg16k/KvaOCz5sAFrIPcYRsPkJ59cHI654//tUxB0ACVxpSa0ZLzFaHuemtLAB1S3FXV3M9RQBr9lKVOfZShAAGAH+2CIhgKBkGaApNFsCW85KcpR2xnjc/Fj/nfHDGcFJokbg7AlDJAYFBmUsLB2PYx+5Z6cbJ5h1VF0xzdj3/98/9f/MdX1x/H+/+Ymo3yoK9tbY4lUPb77/fbVu2767TV3AB0smfyxrE2NvdyRI0S1yQEh1GxvW5DtFzLmU6eOouyHiOD4OQ4DJTMJxaEIj3nhPFZk/DLQ+//tUxCGAET0lf7mHohkqHS1/mDABPZkOi7lDJ2ra71FpXO1ZRwYG1VnW5w5IsDFM03W/o4skN2q3kkuayQn/1/6e80TOpKX980vTNImf/r//3+6f7/7Jt9M4OKzHFRNzFMg7ZYAaLyeuLFB4Fx1rq2yQ4W0bNi6xkX3uZybBiZqrtx3rGiF6K4tXPJzIilZE/hfxPKn9gRdKcz4+juOCThES29Kb0spSGcjEimIGbbAib0BQkbchYgACjsJFeDUY//tUxAyACYRvaewwbMEyDm81l5kugBakupuK5s6mHRoqKCmCa9z3ZxBhEMqhV+nvkIAyzr3YvnYEDoFFi7mOYW+MYJABB970YvW+2wkwIAB2FG374KFAoFL0HhxJTNzOQlp0odmGomNHqhDG9d4eNCEBwFtHj4QPsNpfid2H7Tb10BhU/+XBMfKAW//Q0IBtoeBB4J2gOWVkAl3OcK4zScUBdJIScdSPFoeE8MQCERBSZul+decboNztqCOxAYce//tUxBWACWyvaaekS0EiDa31hg2aqXSp3YeyLv//vVSGlYBhgQwz2m09/qA7TsqKKB0+MIUMoNyMAExtuCu2RuO2DscvK8ZwDt0y+FWvRDqcXoiufHeUcaLkcFMEDuzigI9w3IhZSnmDGBv+qTSJnlwnI1afcRGpfZBwWA3aoAEKAAB9lYEw3w3o5BbFQEpRA9BbhI3h6aL8gJmwd5oIjD6G3FolyZgSWBFOQlUNWZwxoogr//pu6V9kWal91PVq//tUxCEACU0JZaekTQEloK4o8Ypaf7/t+1nBkiU/Q0gEpZZhIJyAYL4mUMQtKLg8W8p1wuXio6Qw8ZsKmtju6h50kKyeJpyGLE8xP//YtSIZl/6tU7yjKjejo/Z3TTWX3QGUiismPqkgUgEEo23BsLwVMxq1MXCYzVNrzk1TAnpOJz/U7VmZT3ynM8I5DHUqs6TCr//231Uqui2ZNKM6gKWrbdHWmjmpojgnIQGxTGtlYg6cbJSSLdkbg56ElSBv//tUxCyACX0Zd6eYTXEmF+809I0mtxXwg45xho50tUNJsFzV/xr0nGpGCJLd3fQVntu6Nnn+edO/fLI1xNmv+x4UciUR5aVWq8Oi4hHBZohDlQQQQBAC5dcBkO480kgm9VH+dTI/ViZZH9VM5n64uKHsrx+RlSqvg4CkDRs86GUJZpk9dP9FJwqETK91VlZr//+kpVHYUWlySTPWXQCkABMUbwuCGI9dG2vqtEp9RKRPKpvnlquwEDEh39ZXNjGx//tUxDcACWD1YaeMUskslKto8xXg8Cyaq9bszFJmdnKzlSpH9/T0iAwJNBgQMNCAm9d/qPHCGQFiAogCTaoASFp2lJptQZI0KOGY783i5l8QldsKyoGQICYDmiwpimgttX45nEhS5qCqpMexHyI+Gth//PSllA5geoWYouBKUOqcSDRa115VKhFRAAjEdCTGigLqt49u2ujFTufYW1MaRedBg4E750by15jua0TfWODdR4MC4ZKNWkLu3MMwYCQb//tUxEEACXyjX6eYbOEmCiv1hg0krNEwQO4eEN1Ud6wgHxA5ihKHgzUAAG0DWUqLQzELoUUZmvlhTbKGsGC80Qh2fIvwn16jOk7S0qRA4FXX3VjOrGBOrJ162pRT7a1RFROzXZkIilNbmtvb60f3UTJVANyuW0uNEkCggqVPglxIQNsBCwLRBoyTXDc9VLUh9dapVXxq/oZjGUMKoKRyllZDUf8paGNpNKz98z0fmugMEGEhnEvyXyuBZZUBS27X//tUxEuACSULVUwwSMEkmau09gi+W3ONgaDXMVeQB8h6YhGKFbKOCIXEZIwAVmZsOys571coKqT2y2P6FVpDR1ReGZ0oTf/Zah1Y0PpogtAwYRAUOpQxLe0tRSkAJjbBACQAJhM0GAwCEALuBWOPEiAATCZ1QmzAsqkW4d+BO/RcdxoR2q4CPMWEAeIR74ciHGdScvjF8pSxqS5f/lZM77oT4RaEspUAOWyOOSMkAKBiP0SsQmNqxtFtRZqMbhFB//tUxFeACWDJW6ekanEqGSVd3QxwOQ5jWyeENSH7v5cl0rlD7ZYk8bW1V1EJSDx55G+vO3fXyuO+P/9uf6m4+iUrWOt/vQRed1tv21u1tlrGYkAASU0whB1HWlCKZx6B1mKhq9ouRLFyLCrHNbNBUljcFM9H4YwNgehLoStquhLC9rhaRKE53uHcfhNy+FgL20Q13Pn7z77UiEHW+UCFvX7XKy5pm9NRNs9FGo7tavhw80i6v///+hbar3N5l+4a//tUxGIACUTFR7WUADoupG03HvICjQLzXpP//////TX//+WesJ4hIMc3qAabLnFlwQUiC0u4PZXVlcIiUQfpwLcMWZokYPlCpwDCBiL0Y9xoQ6duY13Yz9u1UJIalJ27vV/pePuaW0jt9HV0z2Mpdvd9d8QAWRLlILAus8EktgIwy6OfycuTlJvB216k8l8KdbULI+VllY1HQ3lsHWr/nGXmFZtjjK98i0e4AcURv//7IRyf/rulaHMWryGJV2dC//tUxEyACZjDW72DgAlKnuvph5V4DRg4enlCyoDG5G0m4iCxURZV2HFMqmSzRh/IiYk3dylCTfRJEJPZA0YFQhDYJHW12YRdtJ5NhIkgGWl//7VuSIoZ//1Uu3qVku7JMyEOo+z120kBuOMAEspKitBrIq4weNuVN3WdmzITSJZghGQnzPXVMctBWlm2yIc1SJa0zCRzkCI3//+sEMg00u6yXfv7loiO6bhHBJBp/yEmPdWkSxxoBGlJ4HKOADDt//tUxFIACXT5d6ekq/ErHa01hIkyQkWdBFPCcoPbThiQFE+Ihk6D8u5Rud6+I1k21vuykYzAOPDykf//9zHdNf1/12RnkS6qpzNKJmBcP2eyA6A624iGszNbCFaPYVeDJV5bkfcz+eos2DzPS6uKiqEtvNVpoadeRv7FOwylGBBiHsyr/0IznEOqOa1VyV7f0ejJpjqcyCnDW0qgLuQgABIA8RRsBnP2E3g9ajVWn23ca+8VvR2zr6RWkrAAqIhE//tUxFwASUD1baeYrVEiny209IlytkPkCTIyLx+fP35QSGRT2//6LClin7nUtDf3dHvtZl8zlDWo+uFGkARpsALlgcVY5/oBywyMCloLdqUvSLP2STVkLSqqBEiWAMb//iQksVXKVv1qUKAiVf//9HVK9FKpRMv6Go4IUWLAFmGw6dKglbw1AASkRCjxIYCgQoUnQFypTmxewXC6rsRmIy2OzOc3quUDjDa8h5ghYdKatVk/+ximqczVRIf73kkB//tUxGgACYTtWaykTUErGinxlIkwEVcV9IBEoZR8wsNA0Hg4wmejQAA1IWEskSBLjBEo+gjQZHjBP1VCVgX0AsOdg+0y95MDu1Fby4GCei2LLL96KNicXqrk/Sfy/y7SSIY4o1tuv/8hChxg5C3pUjk1ABdlcr2whOEIBprxubBsJ4912A4ewoKSzSVjUQEA5hVFJ/7GLzMvuiISEmld3d+ILcPD3/v+39uOXSIEBB0D8y4YAO8PAAAz39Dx/6AG//tUxHIACWh/Qa3gY2EpGad1tIzks9vjrtb8cRbKKUgABvjxEJZAzhvjWZ2QOgWkEZXKaMM6wTIKaI/QtHtY0O+i4c04OMW4O19bCoOBvISPlKSvqvLO2OfT5RRV0o1FXdaUy8ixH9537ik5Ft9rPvauKaiQ7TyP6uLJGapKxfvVcapr7j4ppWs7i1ObnDY/jGc5//h3jyRNf/sLMrI0sRUGqlQQACYIBZijXbCXRiQ5PuKw9ULGmZmGYXdx5Ac4//tUxHyACZCDUZWBgDozJKp3HvAAuhih9+Hxct5bIjdVa/o+690VLSrXU8i9jzUYPgo8wMLDg8QVGCw8SNYhqmLmER0f5NuL/KqoEaYKFACAp2wh3IFIcJwLbgePyZCQIjCJI4hK9LYqBqtMIsf28n50t0vG/P3v95ZnhQMCZQgcKcWVLMv/p6O1KN/YtOZX3Mruc7WIlqYzqAEuQAgMOqOIo8Km2Or9e547MQl+3Xxl/NW4Niztr2u2+z7++w+J//tUxGUACpD7W1zygBk+H2uphgk5h++KW6OiIrlOwpHNv///SZpjJZv/32eZ3dnRdBasYEaZjnBICsACUUXQ9ChB7LeAxKMS0ImmjusyOTfi6NfyWPrDhZ5zbnQmBqorGedcwtMi6Md5F///tmBhBjtKf//muerqY6YQgAKbO6bZ8/b6YJqkKIL+twF0UKoaZWDZChCEcqFmhHbSNGNVCQQUBLW1pI2JlSrJurDBYJpLdZHb1/9byGjB6lRSKzf0//tUxGiACS0BXywwUkEunWxo8wpbQlrVWd1McjEcXbKuEWgUkAAUIQIEXdFeH/TSKBFUa3U/2KxweOCQgHiMygWQEoLSWRI3TmQFQIdZ5E1uFdqv2uGP//8nsn/+1CM5IXB6Qf/QKpNmhZF8jvB02gHEJBGiT4uLAE0hE2HREMRR6KIh0MozU4ZoHzh6se/9buCFatZpnWFN/OcTViVMVFXa+r2Uq7OZSFK23Tb8lxUlnUPkIz99cSxqQKUwRwg8//tUxHOACWDtZ6ewpYEulmw9hIzoiMpZB6qkHztSaoZbSxmJX3+tRmtXCClIiuyRFkN5pYU031djK7/e1X8iq61+X3YlGbO2Jc6qFUkGM0QuI7kHBQQRDq91AUaZljRJboiEMB9GIcygMtkFw4KC4cOiVokHDbmIkVDKqQ6Tt3coecNnYkLQ7DQipGC5Xchx3/r7v8HdhgVxXiHGN+PvyT3RM5+f/DBruaZUVze1J8VjWS1Wos0mk5wEVDYmLEqb//tUxH2ACVENWUwwRYknoOoVgQ45NJuhe6ytuL3J88lWZ6RzEFZvl0ts/G/pe8T+//7s7870BqV620VsCHnwKCH8gYqmiXRrboyGDfBOCIKULhLCYrAeeLcQ+OOj+jvpatDjPCL0YJjQMxXwDMMzdfnbGP/zXPDX4q+br05xz7QaW3e6D13AvcsTWV/zSUb18kjkaSHEgB4MCEUG4oGRnjp8vy2cCzFY6PHJdFRF4yCa0C0iYJJO7W5rk6I2u9HM//tUxIiASYCDXeekZUkcDKw89I0pVHqqLM/tzWT6G0f1Y1kIZKVGbQ4u3zYAIjA5ZnKtjIgukglmx5ldlYfMu+4DXNMAwawhDQLBeAuZUcozHcdozFD/TQGKFmjny9w5tb1pkhcVcNBZcehVCqi3UPEpYbiKF89KgADWRJa6BfKvHHpiiaW4spI9AcvhJVojKYYgE0TuC2RdX4ZJHqipSXr8ZDjBnTUQqM5DNZH07f2tE0kt30//0pIyJMburGcZ//tUxJSACWBPYeewY4knn2108YobKgfCEEf9JRslDjYK9rNs+Yx7l4Pgi24UbYtq9bb0veNUxo5jIsDk603cWysas97nOx2P//+dg5VRas2///7EkhL0TmeREFIY6B5tCIH8k4uNgj181jZmI2gbCqXBrbnGTSY8nCbks+1lGkx5Nq5xeqgTEhUx9Tq3W4hnN///qcJnIUUVUG3N///c5hrknf7slzBStQQCC42lM20BIaEbkEp3IY+gUIwnD9Sx//tUxJ+ACYRzXewkbOEqoGrphAmxFYGkC6Xm433K0hsfPGRpCY0rww1kWFKMcq///mUVW5xRx1ZnXW/z3tOxxAcQhO4eKjQAoQC+kHUSQNA9EcOw+0SlyAvwpoRvHQd2ElCVuWraa75QLBYADx/40kQGuCME0OJs5Tf//srL+1r9naZf3oUUFTAow8dp2KX9sgAAApEBrY2BoqQfhKR5nAc5tGpK/bVSuKyvcv3TQBgeD07w0MEebFMVB4PRjBo///tUxKmASQUPZ0eMUPkhHOzo9JW3bW///rVlb3KhWdrJT/73FkDox2fpSDL1zgAApxIiRoFDAJIpCiZDOhECfDlyVqOQidg6hstQxVO9MhUFTLq64+AYjzIrDFKq///qUTZaSUMZ9rW/2+6EFgIY5XV9Xcwm5lSH4eoAAWQFIygLAzgpo0JWli8IYm9pVE05Oo0Gk0fHrenJPTtXEpua1oqpIN+stGQlGIpzvO/6nFYw1CyaiwoIPvywqNd2AsIh//tUxLaACVDnbaYYTbEhnOxo8ZYnGHBoCDWsO2ynmLolUtpq/SSDFsJRxFpTPUSo8qijqB4G3b0Z0q03P1Vx+wsj34Hotzz9/Mv+REnQhBJAowgN5hfh8YStQsLlFCACoNgZDIj9DsVcQepzaiszJ6Sds0Maypbso5EPSdv/m+OSmNru9oIdLPGXp/nz4ozqyJqazsVB48rX+nCPz8lJ+nuViZb3J0KE+iJJQggMEobn7AIszEQqw5rWSBkmIfjk//tUxMKACP0BW6wMT2kuoKw08ZYW+gCAlMoX2idd1YqnakRBLSYDnT1k0hmpALC8zW6858sN5oepm4wV7Nypuu/3GwB7is2MDrOBhxYHhQbiaot/vrJZbGlA2CaG3BOM02UTYnDSbUTlhKWzFqFnVlHVECEMozOrnSO+GV7JqToFKRKJEZwO027/c4n/RH1b6rXw/7mjORfIrwdvvfgEsS7uqHLa2BxCGQKKY12E/1hCzBT067OiIQCyByyikpMu//tUxM4ASRS3U0wkqtEdlKnplgyo8bF9L5gxBj1tDh2R14c+e2ZkRH6ycBq0LRAsQzIo6kcGk7ltfmiQvZozJqXnWMozg1B8pVUDepilckckqAgwFEhyQKduAMTDEQwaEwLUq5tIyyqXvX1DKLCRAvSuMERsztUzYtL+Zlw8kVM1mcknqtZLYpkUXnr3Yte3zZ1BwdSM9t2MTB0JSQCku0K5i3Y0AgXglBYKDQgAQaHgJykTg7TKEjjxiwscCB5e//tUxNuACmUJSsyYb8kok6x89gw19IwAmUhqr/RRkLJjIVnfmRzy8c39W8uWolqPkxki/n/bdmvXHgwlhAZqtyI1h0zEpWoAb/Qoq6AFGVdD0GZM0nleT7UHUa/clEHTUsyq27+yBVBtrYQpqxwhcptMmzPpRi/VPnNkmxdLLLd//K/OqvYb6Tpdlyz4xZdJtVImuexq2V1A1NTXJL1rABowolKQMgodoE9GvlBii6a7M6dhUKiD2QpuA+UzyXVh//tUxOIACXxzbaewZxlaIet88w2oIDQJXFM5GnJGo+VPKIUC+EfsWspX5934xHtmfIxEmXU07JB83E6liaVVuTcMKtUyZUBjCkDGmK0fpQA2tbc5syAxZZ0RxFTtRcgIiywEaQQRgEJweepxKDGQcSXDZjnNYVonyZ4ZtHeK3WLOqtqo2VI4f05DxKqyuVhzXVxChhctEqn1AUFaGiIO0gBAIyS3ZAMCATBvwkSBHqa4BFN6vagbSsnm2SzTBj8X//tUxOYACmUFWeewZUFRIer9Fgw4hOBpaFrvHjNSkg2tGvmmdHqvXoKGQGVmR//8umhJ2LJZe1EZ3NaqyX/5FdYGgHw3AAHAaYFv5kIEmSGIBqxxrqgrqQ2LJWajSZqUsaONCIDSOOUByZWRbcmiVxzfZ+r+1j63c13Rv7Nb//8/fdl3ilukYiaPMYufctpkWeHJomBVw5aMEqKu7zABAQVUACAHC3f6TAJOaKOLUzPh4BAVZl8HWXlVQhQExRNj//tUxOeACwUNT4wNMclroqkxkw2pogisoJJnGCK3rgau3xMpsao4lks1l0LM8iHy+9Kf++LEoCJ7uYOxR1Ngvmqv/o/x1yMO1QBD/X2gEATaMQlChGMUDiAXCAxo9MzKdduIQNvsRjNsHTWoGkx0edOa+q3Cz1oJWe+bzkinaZOTdMnWW7dtlMrW6UlUg7MCsYpaMcQ5H1PZOcGV1u3kQVABxyuSffJYBsxz4JAQmq2pMsskcIrPdE6CNtyAJgIm//tUxOOAChzTU6wwZSE8oGj1owmxLJCSiNIY6d7ad25TGCIR3IfmZWZL0UlteQ4YYaK951s6//sUzUgQU9aR3QX8tu9Z3e6VzTrb0grrHI5ZIkUBUmJSHkClCUqtEq14o0gyOSGHjO7gMzu+r1hvt0zKpVwxrz6hVc6cfgOmNQcI1ksiv/+gI7gxQpSCiMvYKvx3bn/9Og5tyFpVfECkAEkb4g4PLiMPsrttQrNxcuGGdtIQRuyvScGThQYtC9Nl//tUxOkAC7zhNk5owwFHGWh1pI1RtbqMcxxbwijHxrDzBJRCN5/3yL/77f1UYvFBxVqgjocBE0HBSZtErdMXS3wXY0dqEAAgieSsUNdd9yzpcS4Lg9zhuTIOoQFQsF2iyynERq7lg0naUuyrNQo/tqLNpGAoijT2ZM+vONuf9p7u77XfMfJjTqektydY0FJWeEz5uOX6KPtX/5BqpagIjai4CRG+GijyiANwQb1IJ1gL4n1Kc5ipcy3JvaRo4GBh//tUxOaACvTtP02YT4lYFuj1gw25R0bRw6tOox0Ali0lFRnhrX2Y/ukcuio936uT6//fy2CghwwR8h1r1xe3u6YpAAEwMVBioMIXq+EQMwaNXdqAuY8zOV6MqVMhswOIUtDWWYiSPrrEraQxuX5a6knrc1pKcFrt4klDXFEJItfqzUlPcylrZeGoiIPBteTMBJZF5LoUKuUsBJhxtuWJlMAoAuCDJc8VcnkyZJ9F9e4F1ABGGo2jhbNlnVd5J24d//tUxOUACiidW6eMVnlIGWahhI3R4ySyFlyU7dJAc2cg6Jk2fop15e4bHItBxrUh+Q+5lHRe/NsvOZNHmBlISLalDiH07MIfzkko0UseJToKKItB5vS1cV5LRVLPneLue0PWHu1yRst8diaUk0EED5KId7yuduo7tbeh9xoRrIKxaNf9rpaYy0MSdrvGSkzCu+agAJaQtxogkAkWl12Hq0a1lap/Kt2562pe/DZFqLE3yRM3JKaosvTkfUIV7lJF//tUxOiACyDRKKw8xYk3GmTY8w4QJFhaLL/eCg4gsCAgQJ1I26VyItrmWZVYc+3m3GQyOZlIrQc+m35ilQUKjKpsjIahmEpaRhCWoi58LD0Zs39lXplcgnJCcGK0FVy1EU5EQahZZZ1Rj1YqH8Z3AsVOoZp5vx6x8ltbD0dglEZZ7po47LIJlZuTZPNbJqlG4ZNNHWlY4purNJQUZSWUiFaItimlJC08o7IUjpGmsXMmqMEIDpK0GNypMoJ+KGmz//tUxOqACvi9Gk2wbwEtFGS08w14EXlD8lKn0wUEuVGna65jv1TmStWic2KmkeWzNmpM0fX5W+Ze72q6CTou7UuhqgZIp0VpIIIIgSYEEoOqWvaJIeqOm17UZuAiiiatGj5xO1EaZz2LegXA96bGc53IdBbWNUt2Ep5ks/KqaRQlGwPskRKN7Zuln5lWNLaXWmU6oGcuXuuWOiz0zXQQhLQKEInolki4JiUbNZULs9BnTnCDAHhHTYlRzAED1h2n\" type=\"audio/mpeg\" />\n",
              "                    Your browser does not support the audio element.\n",
              "                </audio>\n",
              "              "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "<style>audio{visibility: hidden;}</style>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from imutils import face_utils\n",
        "import argparse\n",
        "import imutils\n",
        "from IPython.display import HTML, display\n",
        "video_stream()\n",
        "time.sleep(2.0)\n",
        "from IPython.display import Audio, display\n",
        "\n",
        "s1= '/content/ouvrez_les_yeux.mp3'\n",
        "s2 = '/content/regardez_devant.mp3'\n",
        "s3 = '/content/reposer_vous.mp3'\n",
        "def play_sound(audio_file):\n",
        "    audio = Audio(audio_file, autoplay=True)\n",
        "    display(audio)\n",
        "    display(Javascript(\"document.querySelectorAll('audio').forEach(el => el.style.display = 'none')\"))\n",
        "\n",
        "# perging\n",
        "label_html = 'Capturing...'\n",
        "def stop_sound(audio_file):\n",
        "  display(Audio(audio_file, autoplay=False))\n",
        "closed_eye = 0\n",
        "alarm_playing = False\n",
        "bbox = ''\n",
        "COUNTER1 = 0\n",
        "COUNTER2 = 0\n",
        "COUNTER3 = 0\n",
        "EYE_AR_CONSEC_FRAMES = 6\n",
        "(lStart, lEnd) = face_utils.FACIAL_LANDMARKS_IDXS[\"left_eye\"]\n",
        "(rStart, rEnd) = face_utils.FACIAL_LANDMARKS_IDXS[\"right_eye\"]\n",
        "\n",
        "detector = dlib.get_frontal_face_detector()\n",
        "predictor = dlib.shape_predictor('/content/shape_predictor_68_face_landmarks.dat')\n",
        "while True:\n",
        "    js_reply = video_frame(label_html, bbox)\n",
        "    if not js_reply:\n",
        "        break\n",
        "    img = js_to_image(js_reply[\"img\"])\n",
        "    bbox_array = np.zeros([480, 640, 4], dtype=np.uint8)\n",
        "    gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)\n",
        "\n",
        "    faces = detector(gray,0)\n",
        "    eye_closed = False\n",
        "    if len(faces)==0:\n",
        "      textt = \"no face found \"\n",
        "      play_sound(s2)\n",
        "      time.sleep(3)\n",
        "      stop_sound(s3)\n",
        "      cv2.putText(bbox_array, textt, (10, 20),\n",
        "                    cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 2)\n",
        "\n",
        "\n",
        "    if len(faces)>0:\n",
        "      text = \"{} face(s) found\".format(len(faces))\n",
        "      cv2.putText(bbox_array, text, (10, 20),\n",
        "                    cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 0, 255), 2)\n",
        "\n",
        "    for face in faces:\n",
        "\n",
        "                font = cv2.FONT_HERSHEY_SIMPLEX\n",
        "                font_scale = 0.7\n",
        "                font_color = (255, 255, 255)\n",
        "                line_type = 2\n",
        "                text_position = (10, 30)\n",
        "                landmarks = predictor(gray, face)\n",
        "                landmarks_points = np.array([(p.x, p.y) for p in landmarks.parts()])\n",
        "                image_points = np.array([\n",
        "                    (landmarks_points[30][0], landmarks_points[30][1]),\n",
        "                    (landmarks_points[8][0], landmarks_points[8][1]),\n",
        "                    (landmarks_points[36][0], landmarks_points[36][1]),\n",
        "                    (landmarks_points[45][0], landmarks_points[45][1]),\n",
        "                    (landmarks_points[48][0], landmarks_points[48][1]),\n",
        "                    (landmarks_points[54][0], landmarks_points[54][1])\n",
        "                ], dtype=\"double\")\n",
        "\n",
        "                (x, y, w, h) = (face.left(), face.top(), face.width(), face.height())\n",
        "                bbox_array = cv2.rectangle(bbox_array,(x,y),(x+w,y+h),(255,0,0),2)\n",
        "                for (x, y) in landmarks_points:\n",
        "                  cv2.circle(bbox_array, (x, y), 2, (255,255,255), -1)\n",
        "                left_eye = landmarks_points[36:42]\n",
        "                right_eye = landmarks_points[42:48]\n",
        "                le_x, le_y, le_w, le_h = cv2.boundingRect(np.array([left_eye]))\n",
        "                re_x, re_y, re_w, re_h = cv2.boundingRect(np.array([right_eye]))\n",
        "                leftEyeHull = cv2.convexHull(np.array([left_eye]))\n",
        "                rightEyeHull = cv2.convexHull(np.array([right_eye]))\n",
        "                cv2.drawContours(bbox_array, [leftEyeHull], -1, (0, 255, 0), 2)\n",
        "                cv2.drawContours(bbox_array, [rightEyeHull], -1, (0, 255, 0), 2)\n",
        "                ear = (eye_aspect_ratio(left_eye) + eye_aspect_ratio(right_eye)) / 2.0\n",
        "                mar = mouth_aspect_ratio(landmarks_points[48:68])\n",
        "                text_y = max(y - 10, 20)\n",
        "                cv2.putText(bbox_array, f'EAR: {ear:.2f}', (10,90), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 255), 2)\n",
        "                cv2.putText(bbox_array, f'MAR: {mar:.2f}', (10,110), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 255), 2)\n",
        "                nose_points = [landmarks_points[27], landmarks_points[30], landmarks_points[33]]\n",
        "                nar = nose_aspect_ratio(nose_points)\n",
        "                text = f'Nose Aspect Ratio: {nar:.2f}'\n",
        "                cv2.putText(bbox_array, text,(10,130),cv2.FONT_HERSHEY_SIMPLEX,0.5,(255,0,0),2)\n",
        "                eye_left = landmarks_points[36]\n",
        "                eye_right = landmarks_points[45]\n",
        "                nose_tip = landmarks_points[33]\n",
        "                head_angle = calculate_head_angle(np.array(eye_left), np.array(eye_right), np.array(nose_tip))\n",
        "                cv2.putText(bbox_array, f'Head Angle: {head_angle:.2f}',(10,70), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 255), 2)\n",
        "                size = img.shape\n",
        "                frame_height = img.shape[0]\n",
        "                head_tilt_degree, start_point, end_point, end_point_alt = getHeadTiltAndCoords(size, image_points, frame_height)\n",
        "                cv2.putText(bbox_array, f'Head Tilt: {head_tilt_degree[0]:.2f} degrees', (10, 50), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 255), 2)\n",
        "                cv2.line(bbox_array, start_point, end_point, (0, 255, 0), 2)\n",
        "                if 70 < head_angle < 100:\n",
        "                    cv2.putText(bbox_array, \"good\", (x, y-60), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 2)\n",
        "                    COUNTER2 = 0\n",
        "                else:\n",
        "                    cv2.putText(bbox_array, \"Regardez devant vous !\", (x, y-60), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 2)\n",
        "                    COUNTER2 += 1\n",
        "                    if COUNTER2 >= EYE_AR_CONSEC_FRAMES:\n",
        "                        play_sound(s2)\n",
        "                        COUNTER2 = 0\n",
        "\n",
        "\n",
        "\n",
        "                if ear < 0.2:\n",
        "                            cv2.putText(bbox_array, \"Eyes Closed!\", (x, y - 30), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)\n",
        "                            COUNTER1 += 1\n",
        "                            if COUNTER1 >= EYE_AR_CONSEC_FRAMES:\n",
        "                                play_sound(s1)\n",
        "                                cv2.putText(bbox_array, \"Eyes Closed!\", (x, y - 30), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)\n",
        "                                COUNTER1 = 0\n",
        "                else:\n",
        "                            COUNTER1 = 0\n",
        "\n",
        "                if mar > 0.6:\n",
        "                          play_sound(s3)\n",
        "                          cv2.putText(bbox_array, \"Yawning!\", (x, y - 60), cv2.FONT_HERSHEY_SIMPLEX, 0.7, (0, 0, 255), 2)\n",
        "\n",
        "                for (i, (x, y)) in enumerate(landmarks_points):\n",
        "                    if i == 33:\n",
        "                        image_points[0] = np.array([x, y], dtype='double')\n",
        "                        cv2.circle(bbox_array, (x, y), 1, (0, 255, 0), -1)\n",
        "                        cv2.putText(bbox_array, str(i + 1), (x - 10, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.35, (0, 255, 0), 1)\n",
        "                    elif i == 8:\n",
        "                        image_points[1] = np.array([x, y], dtype='double')\n",
        "                        cv2.circle(bbox_array, (x, y), 1, (0, 255, 0), -1)\n",
        "                        cv2.putText(bbox_array, str(i + 1), (x - 10, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.35, (0, 255, 0), 1)\n",
        "                    elif i == 36:\n",
        "                        image_points[2] = np.array([x, y], dtype='double')\n",
        "                        cv2.circle(bbox_array, (x, y), 1, (0, 255, 0), -1)\n",
        "                        cv2.putText(bbox_array, str(i + 1), (x - 10, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.35, (0, 255, 0), 1)\n",
        "                    elif i == 45:\n",
        "                        image_points[3] = np.array([x, y], dtype='double')\n",
        "                        cv2.circle(bbox_array, (x, y), 1, (0, 255, 0), -1)\n",
        "                        cv2.putText(bbox_array, str(i + 1), (x - 10, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.35, (0, 255, 0), 1)\n",
        "                    elif i == 48:\n",
        "                        image_points[4] = np.array([x, y], dtype='double')\n",
        "                        cv2.circle(bbox_array, (x, y), 1, (0, 255, 0), -1)\n",
        "                        cv2.putText(bbox_array, str(i + 1), (x - 10, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.35, (0, 255, 0), 1)\n",
        "                    elif i == 54:\n",
        "                        image_points[5] = np.array([x, y], dtype='double')\n",
        "                        cv2.circle(bbox_array, (x, y), 1, (0, 255, 0), -1)\n",
        "                        cv2.putText(bbox_array, str(i + 1), (x - 10, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.35, (0, 255, 0), 1)\n",
        "                    else:\n",
        "                        cv2.circle(bbox_array, (x, y), 1, (0, 0, 255), -1)\n",
        "                        cv2.putText(bbox_array, str(i + 1), (x - 10, y - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.35, (0, 0, 255), 1)\n",
        "\n",
        "                for p in image_points:\n",
        "                    cv2.circle(bbox_array, (int(p[0]), int(p[1])), 3, (0, 0, 255), -1)\n",
        "\n",
        "                head_tilt_degree, start_point, end_point, end_point_alt = getHeadTiltAndCoords(size, image_points, frame_height)\n",
        "\n",
        "                cv2.line(bbox_array, start_point, end_point, (255, 0, 0), 2)\n",
        "                cv2.line(bbox_array, start_point, end_point_alt,\n",
        "                 (0, 0, 255), 2)\n",
        "\n",
        "    bbox_array[:, :, 3] = (bbox_array.max(axis=2) > 0).astype(int) * 255\n",
        "    bbox_bytes = bbox_to_bytes(bbox_array)\n",
        "    bbox = bbox_bytes\n"
      ],
      "metadata": {
        "id": "w-O5945c5U9m",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 17
        },
        "outputId": "b40d1027-b2f2-479a-94f0-665c4e7d547c"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "        var video;\n",
              "        var div = null;\n",
              "        var stream;\n",
              "        var captureCanvas;\n",
              "        var imgElement;\n",
              "        var labelElement;\n",
              "\n",
              "        var pendingResolve = null;\n",
              "        var shutdown = false;\n",
              "\n",
              "        function removeDom() {\n",
              "            stream.getVideoTracks()[0].stop();\n",
              "            video.remove();\n",
              "            div.remove();\n",
              "            video = null;\n",
              "            div = null;\n",
              "            stream = null;\n",
              "            imgElement = null;\n",
              "            captureCanvas = null;\n",
              "            labelElement = null;\n",
              "        }\n",
              "\n",
              "        function onAnimationFrame() {\n",
              "            if (!shutdown) {\n",
              "                window.requestAnimationFrame(onAnimationFrame);\n",
              "            }\n",
              "            if (pendingResolve) {\n",
              "                var result = \"\";\n",
              "                if (!shutdown) {\n",
              "                    captureCanvas.getContext('2d').drawImage(video, 0, 0, 640, 480);\n",
              "                    result = captureCanvas.toDataURL('image/jpeg', 0.8);\n",
              "                }\n",
              "                var lp = pendingResolve;\n",
              "                pendingResolve = null;\n",
              "                lp(result);\n",
              "            }\n",
              "        }\n",
              "\n",
              "        async function createDom() {\n",
              "            if (div !== null) {\n",
              "                return stream;\n",
              "            }\n",
              "\n",
              "            div = document.createElement('div');\n",
              "            div.style.border = '2px solid black';\n",
              "            div.style.padding = '3px';\n",
              "            div.style.width = '100%';\n",
              "            div.style.maxWidth = '600px';\n",
              "            document.body.appendChild(div);\n",
              "\n",
              "            const modelOut = document.createElement('div');\n",
              "            modelOut.innerHTML = \"Status:\";\n",
              "            labelElement = document.createElement('span');\n",
              "            labelElement.innerText = 'No data';\n",
              "            labelElement.style.fontWeight = 'bold';\n",
              "            modelOut.appendChild(labelElement);\n",
              "            div.appendChild(modelOut);\n",
              "\n",
              "            video = document.createElement('video');\n",
              "            video.style.display = 'block';\n",
              "            video.width = div.clientWidth - 6;\n",
              "            video.setAttribute('playsinline', '');\n",
              "            video.onclick = () => { shutdown = true; };\n",
              "            stream = await navigator.mediaDevices.getUserMedia({video: { facingMode: \"environment\"}});\n",
              "            div.appendChild(video);\n",
              "\n",
              "            imgElement = document.createElement('img');\n",
              "            imgElement.style.position = 'absolute';\n",
              "            imgElement.style.zIndex = 1;\n",
              "            imgElement.onclick = () => { shutdown = true; };\n",
              "            div.appendChild(imgElement);\n",
              "\n",
              "            const instruction = document.createElement('div');\n",
              "            instruction.innerHTML = 'When finished, click here or on the video to stop this demo';\n",
              "            div.appendChild(instruction);\n",
              "            instruction.onclick = () => { shutdown = true; };\n",
              "\n",
              "            video.srcObject = stream;\n",
              "            await video.play();\n",
              "\n",
              "            captureCanvas = document.createElement('canvas');\n",
              "            captureCanvas.width = 640;\n",
              "            captureCanvas.height = 480;\n",
              "            window.requestAnimationFrame(onAnimationFrame);\n",
              "\n",
              "            return stream;\n",
              "        }\n",
              "\n",
              "        async function stream_frame(label, imgData) {\n",
              "            if (shutdown) {\n",
              "                removeDom();\n",
              "                shutdown = false;\n",
              "                return '';\n",
              "            }\n",
              "\n",
              "            var preCreate = Date.now();\n",
              "            stream = await createDom();\n",
              "\n",
              "            var preShow = Date.now();\n",
              "            if (label != \"\") {\n",
              "                labelElement.innerHTML = label;\n",
              "            }\n",
              "\n",
              "            if (imgData != \"\") {\n",
              "                var videoRect = video.getClientRects()[0];\n",
              "                imgElement.style.top = videoRect.top + \"px\";\n",
              "                imgElement.style.left = videoRect.left + \"px\";\n",
              "                imgElement.style.width = videoRect.width + \"px\";\n",
              "                imgElement.style.height = videoRect.height + \"px\";\n",
              "                imgElement.src = imgData;\n",
              "            }\n",
              "\n",
              "            var preCapture = Date.now();\n",
              "            var result = await new Promise(function(resolve, reject) {\n",
              "                pendingResolve = resolve;\n",
              "            });\n",
              "            shutdown = false;\n",
              "\n",
              "            return {'create': preShow - preCreate, 'show': preCapture - preShow, 'capture': Date.now() - preCapture, 'img': result};\n",
              "        }\n",
              "    "
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import numpy as np\n",
        "import cv2\n",
        "import PIL.Image\n",
        "from base64 import b64encode\n",
        "import io\n",
        "import torch\n",
        "\n",
        "model = torch.hub.load('ultralytics/yolov5', 'yolov5s', pretrained=True)\n",
        "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
        "model.to(device)\n",
        "model.eval()\n",
        "\n",
        "\n",
        "detector = dlib.get_frontal_face_detector()\n",
        "predictor = dlib.shape_predictor('/content/shape_predictor_68_face_landmarks.dat')\n",
        "\n",
        "\n",
        "from IPython.display import Audio, display\n",
        "\n",
        "sound_file = '/content/wake_up.mp3'\n",
        "def play_sound():\n",
        "    display(Audio(sound_file, autoplay=True))\n",
        "def stop_sound():\n",
        "    display(Audio(sound_file, autoplay=False))\n",
        "\n",
        "closed_eye_frame_threshold = 200\n",
        "alarm_playing = False\n",
        "\n",
        "\n",
        "def draw_boxes_and_landmarks(img, faces, landmarks_predictor, results):\n",
        "    for result in results.xyxy[0]:\n",
        "        if result[4] > 0.5:\n",
        "            start_point = (int(result[0]), int(result[1]))\n",
        "            end_point = (int(result[2]), int(result[3]))\n",
        "            color = (255, 0, 0)\n",
        "            cv2.rectangle(img, start_point, end_point, color, 2)\n",
        "            label = f'{model.names[int(result[5])]} {result[4]:.2f}'\n",
        "            cv2.putText(img, label, (start_point[0], start_point[1] - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, color, 2)\n",
        "\n",
        "    for face in faces:\n",
        "        landmarks = landmarks_predictor(img, face)\n",
        "\n",
        "        x, y, w, h = face.left(), face.top(), face.width(), face.height()\n",
        "        cv2.rectangle(img, (x, y), (x + w, y + h), (0, 255, 0), 2)\n",
        "\n",
        "        landmarks_points = np.array([(p.x, p.y) for p in landmarks.parts()])\n",
        "        for (x, y) in landmarks_points:\n",
        "            cv2.circle(img, (x, y), 1, (0, 255, 255), -1)\n",
        "\n",
        "        ear_left = eye_aspect_ratio(landmarks_points[36:42])\n",
        "        ear_right = eye_aspect_ratio(landmarks_points[42:48])\n",
        "        ear = (ear_left + ear_right) / 2.0\n",
        "        mar = mouth_aspect_ratio(landmarks_points[48:68])\n",
        "\n",
        "        cv2.putText(img, f'EAR: {ear:.2f}', (x, y - 20), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 2)\n",
        "        cv2.putText(img, f'MAR: {mar:.2f}', (x, y), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 2)\n",
        "\n",
        "\n",
        "    return img\n",
        "\n",
        "\n",
        "\n",
        "def bbox_to_bytes(bbox_array):\n",
        "    bbox_PIL = PIL.Image.fromarray(bbox_array, 'RGB')\n",
        "    iobuf = io.BytesIO()\n",
        "    bbox_PIL.save(iobuf, format='png')\n",
        "    bbox_bytes = 'data:image/png;base64,{}'.format((str(b64encode(iobuf.getvalue()), 'utf-8')))\n",
        "    return bbox_bytes\n",
        "\n",
        "\n",
        "video_stream()\n",
        "while True:\n",
        "    js_reply = video_frame(label_html, bbox)\n",
        "    if not js_reply:\n",
        "        break\n",
        "\n",
        "    img = js_to_image(js_reply[\"img\"])\n",
        "    gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY)\n",
        "    faces = detector(gray)\n",
        "    results = model(img)\n",
        "\n",
        "    img = draw_boxes_and_landmarks(img, faces, predictor, results)\n",
        "    bbox_array = np.array(img)\n",
        "    bbox = bbox_to_bytes(bbox_array)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 121
        },
        "id": "YHxfIlkjwbSI",
        "outputId": "4f598baa-8c9e-4581-af52-2457d0865047"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Using cache found in /root/.cache/torch/hub/ultralytics_yolov5_master\n",
            "YOLOv5 🚀 v7.0-307-g920c721e Python-3.10.12 torch-2.2.1+cu121 CPU\n",
            "\n",
            "Fusing layers... \n",
            "YOLOv5s summary: 213 layers, 7225885 parameters, 0 gradients, 16.4 GFLOPs\n",
            "Adding AutoShape... \n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.Javascript object>"
            ],
            "application/javascript": [
              "\n",
              "        var video;\n",
              "        var div = null;\n",
              "        var stream;\n",
              "        var captureCanvas;\n",
              "        var imgElement;\n",
              "        var labelElement;\n",
              "\n",
              "        var pendingResolve = null;\n",
              "        var shutdown = false;\n",
              "\n",
              "        function removeDom() {\n",
              "            stream.getVideoTracks()[0].stop();\n",
              "            video.remove();\n",
              "            div.remove();\n",
              "            video = null;\n",
              "            div = null;\n",
              "            stream = null;\n",
              "            imgElement = null;\n",
              "            captureCanvas = null;\n",
              "            labelElement = null;\n",
              "        }\n",
              "\n",
              "        function onAnimationFrame() {\n",
              "            if (!shutdown) {\n",
              "                window.requestAnimationFrame(onAnimationFrame);\n",
              "            }\n",
              "            if (pendingResolve) {\n",
              "                var result = \"\";\n",
              "                if (!shutdown) {\n",
              "                    captureCanvas.getContext('2d').drawImage(video, 0, 0, 640, 480);\n",
              "                    result = captureCanvas.toDataURL('image/jpeg', 0.8);\n",
              "                }\n",
              "                var lp = pendingResolve;\n",
              "                pendingResolve = null;\n",
              "                lp(result);\n",
              "            }\n",
              "        }\n",
              "\n",
              "        async function createDom() {\n",
              "            if (div !== null) {\n",
              "                return stream;\n",
              "            }\n",
              "\n",
              "            div = document.createElement('div');\n",
              "            div.style.border = '2px solid black';\n",
              "            div.style.padding = '3px';\n",
              "            div.style.width = '100%';\n",
              "            div.style.maxWidth = '600px';\n",
              "            document.body.appendChild(div);\n",
              "\n",
              "            const modelOut = document.createElement('div');\n",
              "            modelOut.innerHTML = \"Status:\";\n",
              "            labelElement = document.createElement('span');\n",
              "            labelElement.innerText = 'No data';\n",
              "            labelElement.style.fontWeight = 'bold';\n",
              "            modelOut.appendChild(labelElement);\n",
              "            div.appendChild(modelOut);\n",
              "\n",
              "            video = document.createElement('video');\n",
              "            video.style.display = 'block';\n",
              "            video.width = div.clientWidth - 6;\n",
              "            video.setAttribute('playsinline', '');\n",
              "            video.onclick = () => { shutdown = true; };\n",
              "            stream = await navigator.mediaDevices.getUserMedia({video: { facingMode: \"environment\"}});\n",
              "            div.appendChild(video);\n",
              "\n",
              "            imgElement = document.createElement('img');\n",
              "            imgElement.style.position = 'absolute';\n",
              "            imgElement.style.zIndex = 1;\n",
              "            imgElement.onclick = () => { shutdown = true; };\n",
              "            div.appendChild(imgElement);\n",
              "\n",
              "            const instruction = document.createElement('div');\n",
              "            instruction.innerHTML = 'When finished, click here or on the video to stop this demo';\n",
              "            div.appendChild(instruction);\n",
              "            instruction.onclick = () => { shutdown = true; };\n",
              "\n",
              "            video.srcObject = stream;\n",
              "            await video.play();\n",
              "\n",
              "            captureCanvas = document.createElement('canvas');\n",
              "            captureCanvas.width = 640;\n",
              "            captureCanvas.height = 480;\n",
              "            window.requestAnimationFrame(onAnimationFrame);\n",
              "\n",
              "            return stream;\n",
              "        }\n",
              "\n",
              "        async function stream_frame(label, imgData) {\n",
              "            if (shutdown) {\n",
              "                removeDom();\n",
              "                shutdown = false;\n",
              "                return '';\n",
              "            }\n",
              "\n",
              "            var preCreate = Date.now();\n",
              "            stream = await createDom();\n",
              "\n",
              "            var preShow = Date.now();\n",
              "            if (label != \"\") {\n",
              "                labelElement.innerHTML = label;\n",
              "            }\n",
              "\n",
              "            if (imgData != \"\") {\n",
              "                var videoRect = video.getClientRects()[0];\n",
              "                imgElement.style.top = videoRect.top + \"px\";\n",
              "                imgElement.style.left = videoRect.left + \"px\";\n",
              "                imgElement.style.width = videoRect.width + \"px\";\n",
              "                imgElement.style.height = videoRect.height + \"px\";\n",
              "                imgElement.src = imgData;\n",
              "            }\n",
              "\n",
              "            var preCapture = Date.now();\n",
              "            var result = await new Promise(function(resolve, reject) {\n",
              "                pendingResolve = resolve;\n",
              "            });\n",
              "            shutdown = false;\n",
              "\n",
              "            return {'create': preShow - preCreate, 'show': preCapture - preShow, 'capture': Date.now() - preCapture, 'img': result};\n",
              "        }\n",
              "    "
            ]
          },
          "metadata": {}
        }
      ]
    }
  ]
}