Update with more comments and description

Author: adipandas

tracking.ipynb

@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -13,22 +13,20 @@
    ]
   },
   {
-   "cell_type": "code",
-   "execution_count": 2,
+   "cell_type": "markdown",
    "metadata": {},
-   "outputs": [],
    "source": [
-    "maxLost = 60   # maximum number of lost"
+    "##### Object Tracking Class"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [],
    "source": [
     "class Tracker:\n",
-    "    def __init__(self, maxLost = maxLost):\n",
+    "    def __init__(self, maxLost = 30):           # maxLost: maximum object lost counted when the object is being tracked\n",
     "        self.nextObjectID = 0                   # ID of next object\n",
     "        self.objects = OrderedDict()            # stores ID:Locations\n",
     "        self.lost = OrderedDict()               # stores ID:Lost_count\n",
@@ -41,8 +39,7 @@
     "        \n",
     "        self.nextObjectID += 1\n",
     "    \n",
-    "    def removeObject(self, objectID):\n",
-    "        # remove tracker data after object is lost\n",
+    "    def removeObject(self, objectID):                          # remove tracker data after object is lost\n",
     "        del self.objects[objectID]\n",
     "        del self.lost[objectID]\n",
     "    \n",
@@ -109,28 +106,86 @@
     "        return self.objects\n"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "##### Loading Object Detector Model\n",
+    "\n",
+    "Here, the Face Detection Caffe Model is used.\n",
+    "\n",
+    "The files are taken from the following link:\n",
+    "https://github.com/opencv/opencv_3rdparty/tree/dnn_samples_face_detector_20170830\n"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [],
    "source": [
-    "tracker = Tracker()\n",
-    "caffemodel = {\"prototxt\":\"./deploy.prototxt\", \"model\":\"./res10_300x300_ssd_iter_140000.caffemodel\", \"acc_threshold\":0.50}\n",
+    "caffemodel = {\"prototxt\":\"./deploy.prototxt\",\n",
+    "              \"model\":\"./res10_300x300_ssd_iter_140000.caffemodel\",\n",
+    "              \"acc_threshold\":0.50                  # neglected detections with probability less than acc_threshold value\n",
+    "             }\n",
+    "\n",
     "net = cv.dnn.readNetFromCaffe(caffemodel[\"prototxt\"], caffemodel[\"model\"])"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "##### Instantiate the Tracker Class"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": null,
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "maxLost = 60   # maximum number of object losts counted when the object is being tracked\n",
+    "tracker = Tracker(maxLost = maxLost)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "##### Initiate opencv video capture object\n",
+    "\n",
+    "The `video_src` can take two values:\n",
+    "1. If `video_src=0`: OpenCV accesses the camera connected through USB\n",
+    "2. If `video_src='video_file_path'`: OpenCV will access the video file at the given path (can be MP4, AVI, etc format)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "video_src = 0\n",
+    "cap = cv.VideoCapture(video_src)  "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "##### Start object detection and tracking"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
    "metadata": {
     "scrolled": true
    },
    "outputs": [],
    "source": [
-    "cap = cv.VideoCapture(0)\n",
-    "\n",
-    "(H, W) = (None, None)\n",
+    "(H, W) = (None, None)  # input image height and width for the network\n",
     "\n",
     "while(True):\n",
     "    \n",
@@ -147,9 +202,9 @@
     "    blob = cv.dnn.blobFromImage(image, 1.0, (W, H), (104.0, 177.0, 123.0))\n",
     "    \n",
     "    net.setInput(blob)\n",
-    "    detections = net.forward()\n",
+    "    detections = net.forward()                   # detect objects using object detection model\n",
     "    \n",
-    "    detections_bbox = []   # bounding box for detections\n",
+    "    detections_bbox = []                         # bounding box for detections\n",
     "    \n",
     "    for i in range(0, detections.shape[2]):\n",
     "        if detections[0, 0, i, 2] > caffemodel[\"acc_threshold\"]:\n",
@@ -160,7 +215,7 @@
     "            (startX, startY, endX, endY) = box.astype(\"int\")\n",
     "            cv.rectangle(image, (startX, startY), (endX, endY), (0, 255, 0), 2)    \n",
     "    \n",
-    "    objects = tracker.update(detections_bbox)\n",
+    "    objects = tracker.update(detections_bbox)                  # update tracker based on the newly detected objects\n",
     "    \n",
     "    for (objectID, centroid) in objects.items():\n",
     "        text = \"ID {}\".format(objectID)\n",