predict_video.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import the necessary packages\n",
    "from tensorflow.keras.models import load_model\n",
    "from collections import deque\n",
    "import numpy as np\n",
    "import argparse\n",
    "import pickle\n",
    "import cv2\n",
    "import os\n",
    "from random import randrange\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "img_dir = 'output_images/'\n",
    "model = 'model/activity.model'\n",
    "label_bin = 'model/lb.pickle'\n",
    "input_dir = 'example_clips/example6.mp4'\n",
    "output_dir = 'output/example6_out.mp4'\n",
    "size = 128\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] loading model and label binarizer...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-06-17 18:44:28.530488: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA\n",
      "2022-06-17 18:44:28.548381: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x7f980448f450 executing computations on platform Host. Devices:\n",
      "2022-06-17 18:44:28.548393: I tensorflow/compiler/xla/service/service.cc:175]   StreamExecutor device (0): Host, Default Version\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# load the trained model and label binarizer from disk\n",
    "print(\"[INFO] loading model and label binarizer...\")\n",
    "model = load_model(model)\n",
    "lb = pickle.loads(open(label_bin, \"rb\").read())\n",
    "\n",
    "# initialize the image mean for mean subtraction along with the\n",
    "# predictions queue\n",
    "mean = np.array([123.68, 116.779, 103.939][::1], dtype=\"float32\")\n",
    "Q = deque(maxlen=size)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# initialize the video stream, pointer to output video file, and\n",
    "# frame dimensions\n",
    "vs = cv2.VideoCapture(input_dir)\n",
    "writer = None\n",
    "(W, H) = (None, None)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "OpenCV: FFMPEG: tag 0x47504a4d/'MJPG' is not supported with codec id 7 and format 'mp4 / MP4 (MPEG-4 Part 14)'\n",
      "OpenCV: FFMPEG: fallback to use tag 0x7634706d/'mp4v'\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] cleaning up...\n"
     ]
    },
    {
     "ename": "",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31mThe Kernel crashed while executing code in the the current cell or a previous cell. Please review the code in the cell(s) to identify a possible cause of the failure. Click <a href='https://aka.ms/vscodeJupyterKernelCrash'>here</a> for more info. View Jupyter <a href='command:jupyter.viewOutput'>log</a> for further details."
     ]
    }
   ],
   "source": [
    "\n",
    "# loop over frames from the video file stream\n",
    "while True:\n",
    "    # read the next frame from the file\n",
    "    (grabbed, frame) = vs.read()\n",
    "\n",
    "    # if the frame was not grabbed, then we have reached the end\n",
    "    # of the stream\n",
    "    if not grabbed:\n",
    "        break\n",
    "\n",
    "    # if the frame dimensions are empty, grab them\n",
    "    if W is None or H is None:\n",
    "        (H, W) = frame.shape[:2]\n",
    "\n",
    "    # clone the output frame, then convert it from BGR to RGB\n",
    "    # ordering, resize the frame to a fixed 224x224, and then\n",
    "    # perform mean subtraction\n",
    "    output = frame.copy()\n",
    "    frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)\n",
    "    frame = cv2.resize(frame, (224, 224)).astype(\"float32\")\n",
    "    frame -= mean\n",
    "\n",
    "    # make predictions on the frame and then update the predictions\n",
    "    # queue\n",
    "    preds = model.predict(np.expand_dims(frame, axis=0))[0]\n",
    "    Q.append(preds)\n",
    "\n",
    "    # perform prediction averaging over the current history of\n",
    "    # previous predictions\n",
    "    results = np.array(Q).mean(axis=0)\n",
    "    i = np.argmax(results)\n",
    "    label = lb.classes_[i]\n",
    "\n",
    "    # draw the activity on the output frame\n",
    "    text = \"activity: {}\".format(label)\n",
    "    cv2.putText(output, text, (35, 50), cv2.FONT_HERSHEY_SIMPLEX,\n",
    "                1.25, (255, 0, 0), 5)\n",
    "\n",
    "# saving the predicted video into output folder\n",
    "    # check if the video writer is None\n",
    "    if writer is None:\n",
    "        # initialize our video writer\n",
    "        fourcc = cv2.VideoWriter_fourcc(*\"MJPG\")\n",
    "        writer = cv2.VideoWriter(output_dir, fourcc, 30,\n",
    "                                 (W, H), True)\n",
    "\n",
    "    # write the output frame to disk\n",
    "    writer.write(output)\n",
    "\n",
    "    # show the output image\n",
    "    cv2.imshow(\"Output\", output)\n",
    "    key = cv2.waitKey(1) & 0xFF\n",
    "\n",
    "    # if the `q` key was pressed, break from the loop\n",
    "    if key == ord(\"q\"):\n",
    "        break\n",
    "\n",
    "# release the file pointers\n",
    "\n",
    "print(\"[INFO] cleaning up...\")\n",
    "writer.release()\n",
    "vs.release()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.7.11 ('mytfenv')",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.11"
  },
  "orig_nbformat": 4,
  "vscode": {
   "interpreter": {
    "hash": "8192ee532802352596a806e8fc09d7f17306933f21a3281dcf9579bb1c479069"
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}