Image Stitching

Module: Computer Vision | Difficulty: Intermediate

Homography Estimation

RANSAC for Homography

1. Sample 4 point correspondences
2. Compute homography
3. Count inliers where
4. Keep with most inliers

Multi-Band Blending

where is the Gaussian pyramid and is the Laplacian pyramid.

Cylindrical Projection

import cv2
import numpy as np

def stitch_images(img1, img2):
    orb = cv2.ORB_create(nfeatures=1000)
    kp1, des1 = orb.detectAndCompute(cv2.cvtColor(img1, cv2.COLOR_BGR2GRAY), None)
    kp2, des2 = orb.detectAndCompute(cv2.cvtColor(img2, cv2.COLOR_BGR2GRAY), None)
    
    bf = cv2.BFMatcher(cv2.NORM_HAMMING)
    matches = bf.knnMatch(des1, des2, k=2)
    good = [m for m, n in matches if m.distance < 0.75 * n.distance]
    
    pts1 = np.float32([kp1[m.queryIdx].pt for m in good])
    pts2 = np.float32([kp2[m.trainIdx].pt for m in good])
    H, mask = cv2.findHomography(pts1, pts2, cv2.RANSAC, 5.0)
    
    h1, w1 = img1.shape[:2]
    h2, w2 = img2.shape[:2]
    result = cv2.warpPerspective(img1, H, (w1 + w2, max(h1, h2)))
    result[0:h2, 0:w2] = img2
    return result

Key Takeaways

• RANSAC robustly estimates homography despite outliers
• Multi-band blending creates seamless transitions
• Cylindrical projection handles wide field-of-view panoramas