[BUG] bug_issue_title

[youxiao888]

Hello, when there are two or more capture points in the current image, how can this code ensure that the points on one image correspond one-to-one with the points on another image to solve the fundamental matrix between the two cameras

@socketio.on("calculate-camera-pose")
def calculate_camera_pose(data):
    cameras = Cameras.instance()
    image_points = np.array(data["cameraPoints"])
    image_points_t = image_points.transpose((1, 0, 2))
    camera_poses = [{
        "R": np.eye(3),
        "t": np.array([[0],[0],[0]], dtype=np.float32)
    }]
    for camera_i in range(0, cameras.num_cameras-1):
        camera1_image_points = image_points_t[camera_i]
        camera2_image_points = image_points_t[camera_i+1]
        not_none_indicies = np.where(np.all(camera1_image_points != None, axis=1) & np.all(camera2_image_points != None, axis=1))[0]
        camera1_image_points = np.take(camera1_image_points, not_none_indicies, axis=0).astype(np.float32)
        camera2_image_points = np.take(camera2_image_points, not_none_indicies, axis=0).astype(np.float32)

        F, _ = cv.findFundamentalMat(camera1_image_points, camera2_image_points, cv.FM_RANSAC, 1, 0.99999)
        E = cv.sfm.essentialFromFundamental(F, cameras.get_camera_params(0)["intrinsic_matrix"], cameras.get_camera_params(1)["intrinsic_matrix"])

[jyjblrd]

When collecting data for the camera pose calibration, you must only have one point in the room. If there is only one point seen by each camera, the point association is trivial.