Object Tracking In Motion

Drive Link:  https://drive.google.com/drive/folders/1USa7N0JCIeGGhZhkKP18-QuziOzh8riX?usp=share_link

Code for Single object Tracking:

import cv2

import imutils

TrDict = {'csrt': cv2.TrackerCSRT_create,

         'kcf' : cv2.TrackerKCF_create,

         'boosting' : cv2.TrackerBoosting_create,

         'mil': cv2.TrackerMIL_create,

         'tld': cv2.TrackerTLD_create,

         'medianflow': cv2.TrackerMedianFlow_create,

         'mosse':cv2.TrackerMOSSE_create}

tracker = TrDict['csrt']()

#tracker = cv2.TrackerCSRT_create()

v = cv2.VideoCapture(r'mot.mp4')

ret, frame = v.read()

#frame = imutils.resize(frame,width=600)

cv2.imshow('Frame',frame)

bb = cv2.selectROI('Frame',frame)

tracker.init(frame,bb)

while True:

    ret, frame = v.read()

    if not ret:

        break

    #frame = imutils.resize(frame,width=600)

    (success,box) = tracker.update(frame)

    if success:

        (x,y,w,h) = [int(a) for a in box]

        cv2.rectangle(frame,(x,y),(x+w,y+h),(100,255,0),2)

    cv2.imshow('Frame',frame)

    key = cv2.waitKey(5) & 0xFF

    if key == ord('q'):

        break

v.release()

cv2.destroyAllWindows()

Code for Multiple Object Tracking:

import cv2

import numpy as np

TrDict = {'csrt': cv2.TrackerCSRT_create,

         'kcf' : cv2.TrackerKCF_create,

         'boosting' : cv2.TrackerBoosting_create,

         'mil': cv2.TrackerMIL_create,

         'tld': cv2.TrackerTLD_create,

         'medianflow': cv2.TrackerMedianFlow_create,

         'mosse':cv2.TrackerMOSSE_create}

trackers = cv2.legacy.MultiTracker_create()

v = cv2.VideoCapture(r'mot.mp4')

ret, frame = v.read()

k = 4

for i in range(k):

    cv2.imshow('Frame',frame)

    bbi = cv2.selectROI('Frame',frame)

    tracker_i = TrDict['csrt']()

    trackers.add(cv2.legacy.TrackerCSRT_create(),frame,bbi)

frameNumber = 2

baseDir = r'TrackingResults'

while True:

    ret, frame = v.read()

    if not ret:

        break

    (success,boxes) = trackers.update(frame)

    np.savetxt(baseDir + '/frame_'+str(frameNumber)+'.txt',boxes,fmt='%f')

    frameNumber+=1

    for box in boxes:

        (x,y,w,h) = [int(a) for a in box]

        cv2.rectangle(frame,(x,y),(x+w,y+h),(0,0,255),2)

       

    cv2.imshow('Frame',frame)

    key = cv2.waitKey(5) & 0xFF

    if key == ord('q'):

        break

v.release()

cv2.destroyAllWindows()

boxes = np.loadtxt(r'TrackingResults\frame_102.txt')

boxes