Pixel Correspondence from Disparity Map - OpenCV Python

Question

I am trying to select a point in the left image and get the corresponding point in the right image.
To start with, I did the stereo calibration and rectified the images, and proceeded to find the disparity map. I used the StereoSGBM in OpenCV. Since disparity refers to the distance between two corresponding points in the left and right image of a stereo pair then:

x_right = x_left - Disparity

The problem is that when I implement this it doesn't work. Is the problem my interpretation of the disparity map or its output?

import numpy as np
import cv2 as cv

def mouseCallback(event, x, y, flags, param):
    global mouseX, mouseY
    if event == cv.EVENT_LBUTTONDOWN:
        mouseX = x
        mouseY = y
# Images
img_left = cv2.imread(img_path1)
img_right = cv2.imread(img_path2)
# Grayscale
frame_left = cv.cvtColor(img_left,cv.COLOR_BGR2GRAY)
frame_right = cv.cvtColor(img_right,cv.COLOR_BGR2GRAY)

# Retification
...
# Calculate disparity
cv.namedWindow("Stereo Pair Distance")
while True:
   # SGBM Parameters -----------------
   window_size = 3  
   left_matcher = cv.StereoSGBM_create(
   minDisparity=-16, numDisparities=16+16,  
   blockSize=window_size,
   P1=8 * 2 * window_size**2,
   P2=32 * 2 * window_size**2,
   disp12MaxDiff=0,
   uniquenessRatio=5,
   speckleWindowSize=200,
   speckleRange=2,
   preFilterCap=63,
   mode=cv.STEREO_SGBM_MODE_SGBM_3WAY
  )
    right_matcher = cv.ximgproc.createRightMatcher(left_matcher)
    # FILTER Parameters
    lmbda = 8000
    sigma = 1.4
    # WLS Filter
    wls_filter = cv.ximgproc.createDisparityWLSFilter(matcher_left=left_matcher)
    wls_filter.setLambda(lmbda)
    wls_filter.setSigmaColor(sigma)

    # Compute disparity
    displ = left_matcher.compute(left_gray_image, right_gray_image)  
    dispr = right_matcher.compute(right_gray_image, left_gray_image)  
    displ = np.int16(displ)
    dispr = np.int16(dispr)
    filteredImg = wls_filter.filter(displ, left_gray_image, None, dispr)
    filteredImg = cv.normalize(src=filteredImg, dst=filteredImg, beta=0, alpha=255, norm_type=cv.NORM_MINMAX);
    filteredImg = np.uint8(filteredImg)

    imOut = cv.line(imOut, (640, mouseY),(640*2, mouseY), (0, 0, 255), 2)
    # Left Point
    imOut = cv.circle(imOut, (mouseX,mouseY), 5,(0,255,0), 3)  
    # Right Point
    Right_X = round(mouseX-filteredImg[mouseX,mouseY])
    imOut = cv.circle(imOut, (Right_X+640,mouseY), 5,(0,255,), 3)

    cv.imshow("Stereo Pair Distance",imOut)
    key = cv.waitKey(1)
    if key == ord('q'):
        # Quit when q is pressed
        break
    
cv.destroyAllWindows()

Stereo Pair Rectified

Stereo Correspondence

Disparity Map