(Python)Implement using PnP methods from OpenCV to compute the camera pose. In this example, we have a flat plate on a table. We can use Harris corner detection to detect the four corners of the plate on the image. Then we can apply PnP methods to compute the camera pose with respect to the plate. Here, the 3D coordinates of the four corners are (-0.05, -0.05, 0), (0.05, -0.05, 0), (0.05, 0.05, 0) and (-0.05, 0.05, 0). ******################################## Programming homework.. .. ..#################################..# Programming homework: use PnP methods from opency to compute the camera pose# so far, we have found the four corners on the image# the 3D locations of the four corners are (-0.05, -0.05, 0), (0.05, -0.05, 0), (0.05, 0.05, 0), (-0.05, 0.05, 0)# i.e., we define the world origin in the center of the plate, and the z plane is the plate plane# intrinsic matrix of the cameraintrinsic_matrix = self.projection_to_intrinsics(self._proj_matrix,#Step 1, figure out an assignment of the 3D locations to the detected corners on the image# corners is with shape (4, 2)# The output of step 1 should be a numpy array with shape (4, 3), let's say x3d# Note, there can be multiple possible assignments, find one of themprint('3D locations')# dummy: replace with your solutionx3dprint(x3d)# Step 2: solve the pnp problem using opencv# Refer to this link for the cv2.solve PnP https://www.pythonpool.com/opencv-solvepnp/# The output of step 2 should be a rotation matrix R (3x3) and a translation vector tvec (3x1) of the camera pose# useful functions: cv2.solvePnP, cv2.Rodriguesnp.zeros((4, 3), dtype=np.float32)self._window_width, self._window_height)print('3D rotation')# dummy: replace with your solutionR = np.zeros((3, 3), dtype=np.float32)print(R, R.shape)print('3D translation')# dummy: replace with your solutiontvecprint (tvec, tvec.shape)np.zeros((3, 1), dtype=np. float32)# Step 3: project the 3D points x3d using the estimated camera pose R, tvec (extrinsics) and the camera intrinsics to verify the solution# the output of step 3 should be a numpy array with shape (4, 2), let's say x2dprint('projected 2D locations according to the estimated camera pose')# dummy: replace with your solutionx2dnp.zeros((4, 2), dtype=np. float32)print(x2d, x2d.shape)

Question

Accepted Answer

Given :-
In the above given question,  the statement is mention in the above given question 
 
Need…