OpenCV AttributeError module 'cv2.cv2' has no attribute 'Tracker_create'

Question

I have tried to run this code but get an Attribute Error. Any help would be greatly appreciated.

    import cv2
    import sys

    (major_ver, minor_ver, subminor_ver) = (cv2.__version__).split('.')

    if __name__ == '__main__':

    # Set up tracker.
    # Instead of MIL, you can also use

    tracker_types = ['BOOSTING', 'MIL','KCF', 'TLD', 'MEDIANFLOW', 'CSRT', 'MOSSE']
    tracker_type = tracker_types[5]

    if int(minor_ver) < 3:
        tracker = cv2.cv2.Tracker_create(tracker_type)
    else:
        if tracker_type == 'BOOSTING':
            tracker = cv2.TrackerBoosting_create()
        if tracker_type == 'MIL':
            tracker = cv2.TrackerMIL_create()
        if tracker_type == 'KCF':
            tracker = cv2.TrackerKCF_create()
        if tracker_type == 'TLD':
            tracker = cv2.TrackerTLD_create()
        if tracker_type == 'MEDIANFLOW':
            tracker = cv2.TrackerMedianFlow_create()
        if tracker_type == 'CSRT':
            tracker = cv2.TrackerCSRT_create()
        if tracker_type == 'MOSSE':
            tracker = cv2.TrackerMOSSE_create()

    # Read video
    video = cv2.VideoCapture("./videos/chaplin.mp4")

    # Exit if video not opened.
    if not video.isOpened():
        print("Could not open video")
        sys.exit()

    # Read first frame.
    ok, frame = video.read()
    if not ok:
        print('Cannot read video file')
        sys.exit()

    # Define an initial bounding box
    bbox = (287, 23, 86, 320)

    # Uncomment the line below to select a different bounding box
    bbox = cv2.selectROI(frame, False)

    # Initialize tracker with first frame and bounding box
    ok = tracker.init(frame, bbox)

    while True:
        # Read a new frame
        ok, frame = video.read()
        if not ok:
            break

        # Start timer
        timer = cv2.getTickCount()

        # Update tracker
        ok, bbox = tracker.update(frame)

        # Calculate Frames per second (FPS)
        fps = cv2.getTickFrequency() / (cv2.getTickCount() - timer);

        # Draw bounding box
        if ok:
            # Tracking success
            p1 = (int(bbox[0]), int(bbox[1]))
            p2 = (int(bbox[0] + bbox[2]), int(bbox[1] + bbox[3]))
            cv2.rectangle(frame, p1, p2, (255,0,0), 2, 1)
        else :
            # Tracking failure
            cv2.putText(frame, "Tracking failure detected", (100,80), cv2.FONT_HERSHEY_SIMPLEX, 0.75,(0,0,255),2)

        # Display tracker type on frame
        cv2.putText(frame, tracker_type + " Tracker", (100,20), cv2.FONT_HERSHEY_SIMPLEX, 0.75, (50,170,50),2);

        # Display FPS on frame
        cv2.putText(frame, "FPS : " + str(int(fps)), (100,50), cv2.FONT_HERSHEY_SIMPLEX, 0.75, (50,170,50), 2);

        # Display result
        cv2.imshow("Tracking", frame)

        # Exit if ESC pressed
        k = cv2.waitKey(1) & 0xff
        if k == 27 : break

Output:

Traceback (most recent call last): File "C:\Users\Jatin\OpenCV-Object-Tracking\index.py", line 15, in tracker = cv2.cv2.Tracker_create(tracker_type) AttributeError: module 'cv2.cv2' has no attribute 'Tracker_create'

Solution 1

It depends on which packages of OpenCV and the version you have installed.

I believe you need OpenCV 3.4+ to run those models. Some tracker models are available in 3.2, and 3.3. All trackers in your script are available in 3.4

OpenCV packages: opencv-python: This repository contains the main modules of the OpenCV library.

opencv-contrib-python: The opencv-contrib-python repository contains both the main modules along with the contrib modules

python -m pip install opencv-contrib-python, check to see if you have 3.4+, with pip show opencv .

See how to install opencv for more details

Updates

As @user48956 pointed out opencv v 4.5.x has moved some of these algorithms to cv2.legacy. For example, to access TrackerMOSSE_create function. You would have to get it fromcv2.legacy.TrackerMOSSE_create.

I would recommend keeping up with opencv GitHub as some functions/algorithms will likely move around or be deleted.

Solution 2

If you are using OpenCV 4 or above then the above code may not work.

if int(minor_ver) < 3: line breaks for OpenCV 4

You need to install just opencv contrib for this experiment

My Virtual Env contains

pip install -r requirements.txt

dlib==19.19.0
imutils==0.5.3
numpy==1.18.4
opencv-contrib-python==4.2.0.34
scipy==1.4.1

modified code, compatible with OpenCV 4 version

import cv2
import sys

(major_ver, minor_ver, subminor_ver) = (cv2.__version__).split('.')
print(cv2.__version__)

if __name__ == '__main__':

    # Set up tracker.
    # Instead of MIL, you can also use

    tracker_types = ['BOOSTING', 'MIL','KCF', 'TLD', 'MEDIANFLOW', 'CSRT', 'MOSSE']
    tracker_type = tracker_types[6]

    if int(major_ver) < 4 and int(minor_ver) < 3:
        tracker = cv2.cv2.Tracker_create(tracker_type)
    else:
        if tracker_type == 'BOOSTING':
            tracker = cv2.TrackerBoosting_create()
        if tracker_type == 'MIL':
            tracker = cv2.TrackerMIL_create()
        if tracker_type == 'KCF':
            tracker = cv2.TrackerKCF_create()
        if tracker_type == 'TLD':
            tracker = cv2.TrackerTLD_create()
        if tracker_type == 'MEDIANFLOW':
            tracker = cv2.TrackerMedianFlow_create()
        if tracker_type == 'CSRT':
            tracker = cv2.TrackerCSRT_create()
        if tracker_type == 'MOSSE':
            tracker = cv2.TrackerMOSSE_create()

    # Read video
    video = cv2.VideoCapture(0)

    # Exit if video not opened.
    if not video.isOpened():
        print("Could not open video")
        sys.exit()

    # Read first frame.
    ok, frame = video.read()
    if not ok:
        print('Cannot read video file')
        sys.exit()

    # Define an initial bounding box
    bbox = (287, 23, 86, 320)

    # Uncomment the line below to select a different bounding box
    bbox = cv2.selectROI(frame, False)

    # Initialize tracker with first frame and bounding box
    ok = tracker.init(frame, bbox)

    while True:
        # Read a new frame
        ok, frame = video.read()
        if not ok:
            break

        # Start timer
        timer = cv2.getTickCount()

        # Update tracker
        ok, bbox = tracker.update(frame)

        # Calculate Frames per second (FPS)
        fps = cv2.getTickFrequency() / (cv2.getTickCount() - timer);

        # Draw bounding box
        if ok:
            # Tracking success
            p1 = (int(bbox[0]), int(bbox[1]))
            p2 = (int(bbox[0] + bbox[2]), int(bbox[1] + bbox[3]))
            cv2.rectangle(frame, p1, p2, (255,0,0), 2, 1)
        else :
            # Tracking failure
            cv2.putText(frame, "Tracking failure detected", (100,80), cv2.FONT_HERSHEY_SIMPLEX, 0.75,(0,0,255),2)

        # Display tracker type on frame
        cv2.putText(frame, tracker_type + " Tracker", (100,20), cv2.FONT_HERSHEY_SIMPLEX, 0.75, (50,170,50),2);

        # Display FPS on frame
        cv2.putText(frame, "FPS : " + str(int(fps)), (100,50), cv2.FONT_HERSHEY_SIMPLEX, 0.75, (50,170,50), 2);

        # Display result
        cv2.imshow("Tracking", frame)

        # Exit if ESC pressed
        k = cv2.waitKey(1) & 0xff
        if k == 27 : break

Solution 3

I was getting an error with: tracker = cv2.TrackerMOSSE.create()

even thought I had already installed: opencv-contrib-python

I was able to fix the error by using this instead: tracker = cv2.legacy_TrackerMOSSE.create()

Solution 4

My version of Python is 3.7, which doesn't seem to ignore the code below.

if int(minor_ver) < 3:
        tracker = cv2.cv2.Tracker_create(tracker_type)

I simply deleted it and the module runs.

---

An unrelated but hopefully helpful suggestion. At the end of the code above, I added a `destroyAllWindows() command, which cleans up the viewer:

        if k == 27:
            cv2.destroyAllWindows()
            break

Solution 5

For at least for:

• opencv-contrib-python==4.5.2.52

you need to replace:

cv2.TrackerMOSSE_create()

with...

cv2.legacy.TrackerMOSSE_create()

View more solutions

Comments

[24dinitrophenylhydrazine]

For some reason, the latest opencv-contrib-python version (4.5.1.48) at the time of writing did not work for me. But the version (4.2.0.34) used here works, thanks.

[Burak]

I think you are using IDLE. At the end of the program, the window is closed automatically. IDLE is still interactive after reaching the end of the script.

[mherzog]

I was using Spyder. Is that similar to IDLE in this regard?

[Burak]

I tested with Jupyter Notebook. They all are the same in the sense that python still keeps running. Yours is a good advise, I just wanted to note the reason behind.

[user48956]

Same here: opencv-python==4.5.2.52, opencv-python==4.5.2.52 ... --> cv2.TrackerMOSSE_create is missing

[user48956]

At least for: opencv-python==4.5.2.52, opencv-python==4.5.2.52 ... --> cv2.TrackerMOSSE_create is missing

[Prayson W. Daniel]

Have you tried pip install opencv-contrib-python? The contrib has more features. I have not check 4.5.X yet.

[user48956]

Yes, and with the same version. It is moved to cv2.legacy for some reason I reason. I added a new answer.

[Prayson W. Daniel]

Should we not update this answer to include that?

[titusfx]

In openCV 4.5.5 doesn't exist, cv2.legacy, exist some Trackers but not others like CSRT doesn't show up but DaSiamRPN is available