run-object-detection.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "981ab555",
   "metadata": {},
   "source": [
    "# Run Object Detection"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e4ee307",
   "metadata": {},
   "source": [
    "## Load Train Model From Checkpoint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "eff1af4a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import tensorflow as tf\n",
    "from object_detection.utils import label_map_util\n",
    "from object_detection.utils import config_util"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "5c560154",
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# See Github Issue\n",
    "from object_detection.utils import visualization_utils as viz_utils\n",
    "from object_detection.builders import model_builder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e8b98654",
   "metadata": {},
   "outputs": [],
   "source": [
    "CUSTOM_MODEL_NAME = 'my_ssd_mobnet' \n",
    "PRETRAINED_MODEL_NAME = 'ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8'\n",
    "PRETRAINED_MODEL_URL = 'http://download.tensorflow.org/models/object_detection/tf2/20200711/ssd_mobilenet_v2_fpnlite_320x320_coco17_tpu-8.tar.gz'\n",
    "TF_RECORD_SCRIPT_NAME = 'generate_tfrecord.py'\n",
    "LABEL_MAP_NAME = 'label_map.pbtxt'\n",
    "\n",
    "paths = {\n",
    "    'WORKSPACE_PATH': os.path.join('Tensorflow', 'workspace'),\n",
    "    'SCRIPTS_PATH': os.path.join('Tensorflow','scripts'),\n",
    "    'APIMODEL_PATH': os.path.join('Tensorflow','models'),\n",
    "    'ANNOTATION_PATH': os.path.join('Tensorflow', 'workspace','annotations'),\n",
    "    'IMAGE_PATH': os.path.join('Tensorflow', 'workspace','images'),\n",
    "    'MODEL_PATH': os.path.join('Tensorflow', 'workspace','models'),\n",
    "    'PRETRAINED_MODEL_PATH': os.path.join('Tensorflow', 'workspace','pre-trained-models'),\n",
    "    'CHECKPOINT_PATH': os.path.join('Tensorflow', 'workspace','models',CUSTOM_MODEL_NAME), \n",
    "    'OUTPUT_PATH': os.path.join('Tensorflow', 'workspace','models',CUSTOM_MODEL_NAME, 'export'), \n",
    "    'TFJS_PATH':os.path.join('Tensorflow', 'workspace','models',CUSTOM_MODEL_NAME, 'tfjsexport'), \n",
    "    'TFLITE_PATH':os.path.join('Tensorflow', 'workspace','models',CUSTOM_MODEL_NAME, 'tfliteexport'), \n",
    "    'PROTOC_PATH':os.path.join('Tensorflow','protoc')\n",
    " }\n",
    "\n",
    "files = {\n",
    "    'PIPELINE_CONFIG':os.path.join('Tensorflow', 'workspace','models', CUSTOM_MODEL_NAME, 'pipeline.config'),\n",
    "    'TF_RECORD_SCRIPT': os.path.join(paths['SCRIPTS_PATH'], TF_RECORD_SCRIPT_NAME), \n",
    "    'LABELMAP': os.path.join(paths['ANNOTATION_PATH'], LABEL_MAP_NAME)\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5ffc72a7",
   "metadata": {},
   "source": [
    "## Load latest Checkpoint"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "91532cb1",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-01-02 19:38:42.242238: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-01-02 19:38:42.271653: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-01-02 19:38:42.271800: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-01-02 19:38:42.272499: I tensorflow/core/platform/cpu_feature_guard.cc:151] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations:  AVX2 FMA\n",
      "To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.\n",
      "2022-01-02 19:38:42.272983: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-01-02 19:38:42.273164: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-01-02 19:38:42.273319: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-01-02 19:38:42.718291: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-01-02 19:38:42.718472: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-01-02 19:38:42.718600: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:939] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero\n",
      "2022-01-02 19:38:42.718715: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1525] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 4679 MB memory:  -> device: 0, name: NVIDIA GeForce GTX 1060 6GB, pci bus id: 0000:01:00.0, compute capability: 6.1\n"
     ]
    }
   ],
   "source": [
    "# Load pipeline config and build a detection model\n",
    "configs = config_util.get_configs_from_pipeline_file(files['PIPELINE_CONFIG'])\n",
    "detection_model = model_builder.build(model_config=configs['model'], is_training=False)\n",
    "\n",
    "# Restore checkpoint\n",
    "ckpt = tf.compat.v2.train.Checkpoint(model=detection_model)\n",
    "# Select Checkpoint from `TFODCourse/Tensorflow/workspace/models/my_ssd_mobnet`\n",
    "ckpt.restore(os.path.join(paths['CHECKPOINT_PATH'], 'ckpt-3')).expect_partial()\n",
    "\n",
    "@tf.function\n",
    "def detect_fn(image):\n",
    "    image, shapes = detection_model.preprocess(image)\n",
    "    prediction_dict = detection_model.predict(image, shapes)\n",
    "    detections = detection_model.postprocess(prediction_dict, shapes)\n",
    "    return detections"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b5469a8d",
   "metadata": {},
   "source": [
    "## Detection from Image File"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "5b6062b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "import cv2 \n",
    "import numpy as np\n",
    "from matplotlib import pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "655c3986",
   "metadata": {},
   "outputs": [],
   "source": [
    "category_index = label_map_util.create_category_index_from_labelmap(files['LABELMAP'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "bf699fd1",
   "metadata": {},
   "outputs": [],
   "source": [
    "IMAGE_PATH = os.path.join(paths['IMAGE_PATH'], 'test', 'thumbsdown.sdfe6640-fhyg3-hfgec-b6d3-1c1b0dc5817f.jpg')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "883ea58b",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-01-02 19:39:27.920256: I tensorflow/stream_executor/cuda/cuda_dnn.cc:366] Loaded cuDNN version 8300\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "img = cv2.imread(IMAGE_PATH)\n",
    "image_np = np.array(img)\n",
    "\n",
    "input_tensor = tf.convert_to_tensor(np.expand_dims(image_np, 0), dtype=tf.float32)\n",
    "detections = detect_fn(input_tensor)\n",
    "\n",
    "num_detections = int(detections.pop('num_detections'))\n",
    "detections = {key: value[0, :num_detections].numpy()\n",
    "              for key, value in detections.items()}\n",
    "detections['num_detections'] = num_detections\n",
    "\n",
    "# detection_classes should be ints.\n",
    "detections['detection_classes'] = detections['detection_classes'].astype(np.int64)\n",
    "\n",
    "label_id_offset = 1\n",
    "image_np_with_detections = image_np.copy()\n",
    "\n",
    "viz_utils.visualize_boxes_and_labels_on_image_array(\n",
    "            image_np_with_detections,\n",
    "            detections['detection_boxes'],\n",
    "            detections['detection_classes']+label_id_offset,\n",
    "            detections['detection_scores'],\n",
    "            category_index,\n",
    "            use_normalized_coordinates=True,\n",
    "            max_boxes_to_draw=5,\n",
    "            min_score_thresh=.8,\n",
    "            agnostic_mode=False)\n",
    "\n",
    "plt.imshow(cv2.cvtColor(image_np_with_detections, cv2.COLOR_BGR2RGB))\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32e9e9b7",
   "metadata": {},
   "source": [
    "## Real Time Detections from your IP Camera"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "92d434b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# !pip install opencv-contrib-python\n",
    "# !pip uninstall opencv-python-headless -y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "a5a7bd39",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n",
      "QObject::moveToThread: Current thread (0x560b19ff0f20) is not the object's thread (0x560b1a6cad40).\n",
      "Cannot move to target thread (0x560b19ff0f20)\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[h264 @ 0x560b1a537440] error while decoding MB 24 56, bytestream -8\n",
      "[h264 @ 0x560b1aa31180] error while decoding MB 7 44, bytestream -5\n",
      "[h264 @ 0x560b1a80c5c0] cabac decode of qscale diff failed at 45 40\n",
      "[h264 @ 0x560b1a80c5c0] error while decoding MB 45 40, bytestream -6\n",
      "[h264 @ 0x560b1afcfd00] error while decoding MB 68 32, bytestream -6\n",
      "[h264 @ 0x560b1a80c5c0] error while decoding MB 10 17, bytestream -6\n",
      "[h264 @ 0x560b1a537440] error while decoding MB 1 45, bytestream -8\n",
      "[h264 @ 0x560b1ad8bc00] error while decoding MB 34 36, bytestream -6\n",
      "[h264 @ 0x560b1ad8bc00] cabac decode of qscale diff failed at 10 54\n",
      "[h264 @ 0x560b1ad8bc00] error while decoding MB 10 54, bytestream -6\n",
      "[h264 @ 0x560b1aa31180] error while decoding MB 9 52, bytestream -9\n",
      "[h264 @ 0x560b1aab3800] error while decoding MB 25 38, bytestream -7\n",
      "[h264 @ 0x560b1a537440] error while decoding MB 26 10, bytestream -6\n",
      "[h264 @ 0x560b1a85ec40] error while decoding MB 1 59, bytestream -6\n",
      "[h264 @ 0x560b1aff4b40] error while decoding MB 24 50, bytestream -8\n",
      "[h264 @ 0x560b1a80c5c0] error while decoding MB 0 51, bytestream -6\n",
      "[h264 @ 0x560b1afcfd00] error while decoding MB 43 25, bytestream -6\n",
      "[h264 @ 0x560b1aab3800] error while decoding MB 9 51, bytestream -6\n"
     ]
    }
   ],
   "source": [
    "RTSP_URL = 'rtsp://admin:instar@192.168.2.19/livestream/12'\n",
    "os.environ['OPENCV_FFMPEG_CAPTURE_OPTIONS'] = 'rtsp_transport;udp'\n",
    "cap = cv2.VideoCapture(RTSP_URL, cv2.CAP_FFMPEG)\n",
    "\n",
    "width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))\n",
    "height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))\n",
    "\n",
    "while cap.isOpened(): \n",
    "    ret, frame = cap.read()\n",
    "    image_np = np.array(frame)\n",
    "    \n",
    "    input_tensor = tf.convert_to_tensor(np.expand_dims(image_np, 0), dtype=tf.float32)\n",
    "    detections = detect_fn(input_tensor)\n",
    "    \n",
    "    num_detections = int(detections.pop('num_detections'))\n",
    "    detections = {key: value[0, :num_detections].numpy()\n",
    "                  for key, value in detections.items()}\n",
    "    detections['num_detections'] = num_detections\n",
    "\n",
    "    # detection_classes should be ints.\n",
    "    detections['detection_classes'] = detections['detection_classes'].astype(np.int64)\n",
    "\n",
    "    label_id_offset = 1\n",
    "    image_np_with_detections = image_np.copy()\n",
    "\n",
    "    viz_utils.visualize_boxes_and_labels_on_image_array(\n",
    "                image_np_with_detections,\n",
    "                detections['detection_boxes'],\n",
    "                detections['detection_classes']+label_id_offset,\n",
    "                detections['detection_scores'],\n",
    "                category_index,\n",
    "                use_normalized_coordinates=True,\n",
    "                max_boxes_to_draw=5,\n",
    "                min_score_thresh=.8,\n",
    "                agnostic_mode=False)\n",
    "\n",
    "    cv2.imshow('object detection',  cv2.resize(image_np_with_detections, (800, 600)))\n",
    "    \n",
    "    if cv2.waitKey(10) & 0xFF == ord('q'):\n",
    "        cap.release()\n",
    "        cv2.destroyAllWindows()\n",
    "        break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "21632ae3",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "tfodj",
   "language": "python",
   "name": "tfodj"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}