Sei sulla pagina 1di 1

DIP

Assignment 1

General Instructions:

Please form groups of 3. Larger groups will not be acceptable unless there is an exceptional
reason.

For implementation, the suggested programming platforms are MATLAB or python whichever
you feel comfortable. Acknowledge or cite any libraries or code resources you use. We would be
using plagiarism checker for code. Any sort of plagiarism may result in heavy penalization.
Maintain proper indentation and commenting in the code. Also submit a readme file for our ease
to run the code at the time of demos.

You will be required to submit a report (in IEEE


templates http://www.ieee.org/conferences_events/conferences/publishing/templates.html )
stating objective, methodology, results, discussion and conclusion clearly. The report file will be
needed to be uploaded on Turnitin forum.

Code is needed to be uploaded on the Piazza forum in Z_Z_Z(i).zip format where each Z would
be the student name and i is the group number. Also include the .pdf file for the report in the .zip
folder. Clearly mention the group number and group members name and enrollment numbers to

avoid any kind of discrepancy.


There would be no change in the deadline date so try to stick to the deadline.

Deadline: 21st Feb, 2016(Sunday), 12:00:00 a.m.

Assignment 1:
Panorama Stitching
Panorama stitching refers to the seamless composition of multiple images with
overlapped fields of view. Click a set of images for Bharti building from same position
with partial overlaps.
1) Feature extraction: Create your own implementation for the Harris corner detector
and use the existing OpenCV or MATLAB implementations for the SIFT detector for
feature extraction.
2)

Point Correspondences: Find feature point correspondences between images

using homography.
3)

Do image warping. Stitch the images together.

Analysis: Results of image stitching. Also comment on the quality of point


correspondences obtained by your Harris detector and SIFT.

Potrebbero piacerti anche