Loading autocnet/matcher/outlier_detector.py +0 −30 Original line number Diff line number Diff line Loading @@ -150,33 +150,3 @@ def compute_homography(kp1, kp2, outlier_algorithm=cv2.RANSAC, reproj_threshold= return transformation_matrix, mask def homography_test(kp1, kp2, homography, threshold=3.0): """ Utilize the transformation matrix (homography) to check whether keypoint one (`kp1`) to keypoint two (`kp2`). If the point is within threshold units (where the unit is agnositic and a function of the image pixel size) return true. Parameters ---------- kp1 : list of x, y coordinates kp2 : list of x, y coordinates homography : ndarray 3x3 transformation matrix threshold : float The threshold within which true is returned Returns ------- : bool True if within the threshold, else False """ kp1 = kp1[::-1] kp2 = kp2[::-1] kp1 = np.array([*kp1, 1]) kp2 = np.array([*kp2, 1]) print(kp1, np.dot(kp1, homography), kp2) Loading
autocnet/matcher/outlier_detector.py +0 −30 Original line number Diff line number Diff line Loading @@ -150,33 +150,3 @@ def compute_homography(kp1, kp2, outlier_algorithm=cv2.RANSAC, reproj_threshold= return transformation_matrix, mask def homography_test(kp1, kp2, homography, threshold=3.0): """ Utilize the transformation matrix (homography) to check whether keypoint one (`kp1`) to keypoint two (`kp2`). If the point is within threshold units (where the unit is agnositic and a function of the image pixel size) return true. Parameters ---------- kp1 : list of x, y coordinates kp2 : list of x, y coordinates homography : ndarray 3x3 transformation matrix threshold : float The threshold within which true is returned Returns ------- : bool True if within the threshold, else False """ kp1 = kp1[::-1] kp2 = kp2[::-1] kp1 = np.array([*kp1, 1]) kp2 = np.array([*kp2, 1]) print(kp1, np.dot(kp1, homography), kp2)
