Introduction to Computer Vision
Coursework and Assessment
Computer Vision作业代写 The objective of Submission 2 is to compare / apply / design XAI techniques for Computer Vision (X-CV) for black-box CV models.
ECS709: assessment
- Final exam:
50% of the final mark
- one written paper
- Coursework:
50% of the final mark
- individual coursework
- submission 1: five exercises
25% of the final mark
- time available: lab time + individual study hours
- submission: code + report
- evaluated for the quality of the analysis and the discussion of the results obtained Computer Vision作业代写
- submission 2: one short report
25% of the final mark
- time available: individual study hours
- submission: short report (+ supporting material)
- evaluated for the quality of the analysis and the discussion
Submission 1 Computer Vision作业代写
Five exercises
- Geometric Transformations
- Convolution
- Video Segmentation
- Texture Classification
- Object Counting
Report: discussion
- Understanding
- Analysis
- Challenges
- Mistakes
- Discoveries
Submission 2: topic
- Explainable Computer Vision (X-CV)
- Explainable AI (XAI) allows a user (or different types of users) to understand / interpret the predictions of a model (expected impact, potential biases)
- The objective of Submission 2 is to compare / apply / design XAI techniques for Computer Vision (X-CV) for black-box CV models.
- You can choose one of these 3 mini-projects:
- Critical literature review on X-CV (and/or XAI)
- Experimental comparison of existing X-CV methods (or XAI methods applied to CV)
- Your idea for X-CV (with some experimental validation)
Choice 1 – Critical literature review on X-CV (and/or XAI)
Critical review of a set of articles
– discuss key concepts
– assert your perspective(s)
– offer a reflection / critique
– critically evaluate the material
– compare and contrast approaches
– discuss the main ideas / a specific concept / a theme
– demonstrate understanding of the specific topic + its context
… much more than just a summary!
Where to search for the articles?
https://ieeexplore.ieee.org/Xplore/home.jsp
https://scholar.google.com/
https://ieeexplore.ieee.org/document/9369420 (review article)
Choice 2 – Experimental comparison of existing X-CV methods Computer Vision作业代写
(or XAI methods applied to CV)
Choice 3 – Your idea for X-CV (with some experimental validation)
– Select a common objective and a number of baselines for the comparison (i.e. method(s) to compare)
– (Briefly) describe the methods (more in details for yours, in case of Choice 3)
– Select dataset(s) to use for the comparison
– Select performance measure(s) for the comparison
– Present and critically discuss the results Computer Vision作业代写
– Draw conclusions based on the results (and the various choices you made)
Where to search for the implementation of methods?
https://paperswithcode.com
https://github.com
Submission 2
25% of the final mark
- Short report + supporting material
- Supporting material: folder with [as appropriate]
- PDFs of (annotated) references, webpage links
- Code [if applicable]
- Results (e.g. videos, images) [if applicable]Evaluation of Submission 2
- Well-structured, balanced and complete 2-page report: up to 7 marks
- A full and detailed discussion of the results OR literature
+ fully working software OR detailed analysis OR understanding of the key
aspects and limitation of the literature: up to 18 marks
Reference and Lectures Computer Vision作业代写
w.Samek, G. Montavon, S. Lapuschkin, C. J. Anders and K. -R. Müller, “Explaining Deep Neural
Networks and Beyond: A Review of Methods and Applications,” in Proceedings of the IEEE, vol.
109, no. 3, pp. 247-278, March 2021, doi: 10.1109/JPROC.2021.3060483.
https://ieeexplore.ieee.org/document/9369420
2022 Intelligent Sensing Winter School on Explainable AI Sensing (December, 12, 13, 14, 19)
http://cis.eecs.qmul.ac.uk/school2022.html
You can present your idea / mini-project on Dec, 19 as part of the Winter School programme!
更多代写:Final季网课代考 gre作弊 Chemistry网课代修 Essay代写教育学 Formal Writing代写 论文撰写