OpenCV Python Meanshift & Camshift Tutorial: Object Tracking in Video Explained

OpenCV Python – Meanshift and Camshift

Meanshift and Camshift are powerful object tracking algorithms used in computer vision. They help track moving objects in video by following color distribution and object location over time.

In OpenCV Python, these methods are widely used for real-time tracking applications.

---

1. What are Meanshift and Camshift?

Meanshift

Meanshift is a tracking algorithm that:

• Finds the object in a probability distribution
• Moves a window toward the highest density area
• Tracks based on color histogram

Camshift (Continuously Adaptive Meanshift)

Camshift improves Meanshift by:

• Adapting window size automatically
• Handling object scale changes
• Improving tracking accuracy

---

2. Import OpenCV and NumPy

import cv2
import numpy as np

---

3. Read Video

cap = cv2.VideoCapture("video.mp4")

---

4. Select ROI (Region of Interest)

ret, frame = cap.read()
x, y, w, h = cv2.selectROI(frame, False)
track_window = (x, y, w, h)

---

5. Set Up Histogram for Tracking

roi = frame[y:y+h, x:x+w]
hsv_roi = cv2.cvtColor(roi, cv2.COLOR_BGR2HSV)

mask = cv2.inRange(hsv_roi, (0, 60, 32), (180, 255, 255))
roi_hist = cv2.calcHist([hsv_roi], [0], mask, [180], [0, 180])
cv2.normalize(roi_hist, roi_hist, 0, 255, cv2.NORM_MINMAX)

---

6. Meanshift Tracking

term_crit = (cv2.TERM_CRITERIA_EPS | cv2.TERM_CRITERIA_COUNT, 10, 1)

while True:
    ret, frame = cap.read()
    if not ret:
        break

    hsv = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
    dst = cv2.calcBackProject([hsv], [0], roi_hist, [0, 180], 1)

    ret, track_window = cv2.meanShift(dst, track_window, term_crit)

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

    cv2.imshow("Meanshift Tracking", img)

    if cv2.waitKey(30) & 0xFF == 27:
        break

---

7. Camshift Tracking

while True:
    ret, frame = cap.read()
    if not ret:
        break

    hsv = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
    dst = cv2.calcBackProject([hsv], [0], roi_hist, [0, 180], 1)

    ret, track_window = cv2.CamShift(dst, track_window, term_crit)

    pts = cv2.boxPoints(ret)
    pts = np.int32(pts)

    img = cv2.polylines(frame, [pts], True, (0, 255, 0), 2)

    cv2.imshow("Camshift Tracking", img)

    if cv2.waitKey(30) & 0xFF == 27:
        break

---

8. Difference Between Meanshift and Camshift

Feature	Meanshift	Camshift
Window Size	Fixed	Adaptive
Accuracy	Moderate	High
Object Scale Change	No	Yes
Performance	Fast	Slightly slower

---

9. Applications

Meanshift and Camshift are used in:

• Object tracking systems
• Video surveillance
• Motion analysis
• Robotics vision
• Gesture recognition

---

10. Common Mistakes

❌ Wrong ROI selection

✔ Solution:

• Select object carefully using cv2.selectROI()

---

❌ Poor tracking in complex background

✔ Solution:

• Improve lighting and color separation

---

11. Conclusion

Meanshift and Camshift in OpenCV Python are powerful tracking algorithms for real-time object tracking. Camshift improves upon Meanshift by adapting to object size and movement.

Once you master these, you can move to advanced tracking methods like optical flow and deep learning-based tracking.