ECE 564 Computer Vision Assignment 4

Submit HTML documentation on Isidore. Include MATLAB files needed to reproduce your published MATLAB.

  1. Scan the faces of some boxed product and generate and accurate model of that box. Your published MATLAB should include views showing all sides of the box. Bring the box to show in class or send me a photo to ensure there are no duplicates. Save the box for use later in the semester. Make the dimensions match those of the original box. Note that for successful texture mapping, the size of the texture images must not be too large. See Raisin Bran box example.

  2. Use the images taken in class 13 Feb 2017 (available from Isidore resources as CV2017-02-13.zip) and the MATLAB cameraCalibrator to determine the camera intrinsics and the location of the camera in the room (from the Extrinsics for the target boards on the four different walls). Compare the intrinsics to those calculated in assignment 3. Compare the extrinsics distances to those measured directly. See example for a crucial formula.

  3. Using the image 2017-02-13_18-39-28.JPG, find the vanishing point defined by the top and bottom rows and by the left and right columns. The columns look almost parallel to me, but the rows clearly converge. If the vanishing point is not too distant, show the vanishing point on the same figure as the image.

  4. The Bouguet calibrator function rodrigues.m has some hidden code to test the function. Extract this hidden code and use it as part of a test function using published MATLAB principles to demonstrate the use and accuracy of the function.

  5. Calculate the focal length of the MATLAB camera from images from hall model4, which are stored in results.zip. Use this data to determine the numerical value of focal length used in the MATLAB script.

  6. Calibrate the MATLAB camera using the calibration toolbox with images from hall model4, which are stored in results2.zip. Note that the distortion parameters should be zero. Compare the focal length that obtained in the previous problem. Also compare the target locations (from extrinsics) to the MATLAB values.


Maintained by John Loomis, last updated 14 February 2017