stiching.py

import cv2
import numpy as np

def get_sift_homography(img1, img2):

	# Initialize SIFT 
	sift = cv2.xfeatures2d.SIFT_create()

	# Extract keypoints and descriptors
	k1, d1 = sift.detectAndCompute(img1, None)
	k2, d2 = sift.detectAndCompute(img2, None)

	# Bruteforce matcher on the descriptors
	bf = cv2.BFMatcher()
	matches = bf.knnMatch(d1,d2, k=2)

	# Make sure that the matches are good
	verify_ratio = 0.8 # Source: stackoverflow
	verified_matches = []
	for m1,m2 in matches:
		# Add to array only if it's a good match
		if m1.distance < verify_ratio * m2.distance:
			verified_matches.append(m1)

	# Mimnum number of matches
	min_matches = 8
	if len(verified_matches) > min_matches:
		
		# Array to store matching points
		img1_pts = []
		img2_pts = []

		# Add matching points to array
		for match in verified_matches:
			img1_pts.append(k1[match.queryIdx].pt)
			img2_pts.append(k2[match.trainIdx].pt)
		img1_pts = np.float32(img1_pts).reshape(-1,1,2)
		img2_pts = np.float32(img2_pts).reshape(-1,1,2)
		
		# Compute homography matrix
		M, _ = cv2.findHomography(img1_pts, img2_pts, cv2.RANSAC, 5.0)
		return M
	else:
		print ('Error: Not enough matches')
		exit()

img1 = cv2.imread("C:\\Users\\THINK\\Pictures\\house1.jpg")
img2 = cv2.imread("C:\\Users\\THINK\\Pictures\\house2.jpg")

M = get_sift_homography(img1, img2)
result = cv2.warpPerspective(img2,M,(img1.shape[1],img1.shape[0]))
# show the images
img = np.hstack((img1,img2))
cv2.imshow("Image", img)
cv2.imshow("Result", result)
cv2.waitKey(0)