COMP338 – Computer Vision
COMP338 – Computer Vision – Assignment 1
o This assignment is worth 15% of the total mark for COMP338
o Students will do the assignment individually.
Submission Instructions
o Send all solutions as a single PDF document containing your answers, results, anddiscussion of the results. Attach the source code for the programming problems as
separate files. (One PDF doc, one source code file {python or Jupyter NotebookIpython)})
o Each student will make a single submission to the Canvas system.
o The deadline for this assignment 14/11/2024, 5:00pm
o Penalties for late submission apply in accordance with departmental policy as set
out in the student handbook, which can be found at
http://intranet.csc.liv.ac.uk/student/msc-handbook.pdf
and the University Code of Practice on Assessment, found at
https://www.liverpool.ac.uk/media/livacuk/tqsd/code-of-practice-on
assessment/code_of_practice_on_assessment.pdfTask 1. (50 marks) Canny Edge DetectioOpenCV provides a function canny() to get the edge detection result with n image (you canuse any grey image). Please do the following:
- (25 marks) Reimplement the canny operation without using the built-in canny()function (with some explanations of the code).
- (10 marks) Test and visualize your implementation results. (with different filters,different thresholds and others)
- (15 marks) Discuss the difference between your implementation, your resultscompared with the OpenCV implementation.代 写COMP338 – Computer Vision (Compare the numerical results and therunning time and others.)Note:- It is acceptable if the implementations do not match exactly; you will need to analyzethe differences between your implementation and the Canny method. Including yourown reflections in the report will result in additional bonus points. However, it imandatory that you reimplement the function based on your understanding.
Task 2. (50 marks) Feature Extraction
In Lecture 11 and Lab 04 - SIFT & Feature Matching, we have discussed the SIFT feature.In practice, there are several other feature extraction methods such as SURF or ORB. In thistask, we will do extra reading, implementation, and compare SIFT vs. SURF vs. ORB.Papers to read:
- Bay et al., SURF: Speeded Up Robust Features, ECCV 2006
- Rublee et al., ORB: An efficient alternative to SIFT or SURF, ICCV 2011.
Good tutorials:
- (20 marks) Read the SURF and ORB papers and tutorials, summarize yourunderstanding. Compare the differences among SIFT vs. SURF vs. ORB.(10 marks) Given two images (victoria.jpg and victoria2.jpg – both available onCanvas), call OpenCV functions to extract ORB keypoints. You can use the built-infunctions from OpenCV. Visualize the detected keypoints.
- (20 marks) Given two images (victoria.jpg and victoria2.jpg), extract the descriptorsusing SIFT and ORB. Perform keypoint matching using Brute-Force Matcher. Fromthe results, which method do you think perform the best? Justify your answer.Note:- You can also choose the images yourself, as long as they are of the same subject takenfrom different perspectives. You may directly use greyscale images.- Including your own reflections in the report will result in additional bonus points.However, it is mandatory that you reimplement the function based on yourunderstanding.
