In order to be able to use the SIFT functions with python-opencv, we first need to install opencv-contrib by following the instructions from this link: https://pypi.org/project/opencv-contrib-python/. The next code block demonstrates how to detect the SIFT keypoints and draws them using the input Mona Lisa image. 

We will first construct a SIFT object and then use the detect() method to compute the keypoints in an image. Every keypoint is a special feature, and has several attributes. For example, its (x, y) coordinates, angle (orientation), response (strength of keypoints), size of the meaningful neighborhood, and so on.

We will then use the drawKeyPoints() function from cv2 to draw the small circles around the detected keypoints. If the cv2.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS flag is applied to the function, it will draw a circle with the size of a keypoint, along with its orientation. In order to compute the keypoints and the descriptor together, we will use the function detectAndCompute():

# make sure the opencv version is 3.3.0 with
# pip install opencv-python==3.3.0.10 opencv-contrib-python==3.3.0.10

import cv2
print(cv2.__version__)
# 3.3.0

img = cv2.imread('../images/monalisa.jpg')
gray= cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
 
sift = cv2.xfeatures2d.SIFT_create()
kp = sift.detect(gray,None) # detect SIFT keypoints
 
img = cv2.drawKeypoints(img,kp, None, flags=cv2.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS)
cv2.imshow("Image", img);
cv2.imwrite('me5_keypoints.jpg',img)

kp, des = sift.detectAndCompute(gray,None) # compute the SIFT descriptor

Here is the output of the code, the input Mona Lisa image, along with the computed SIFT keypoints drawn on it, and with the orientations: