This paper review is only for degree students.
Due: 23:00 on the 1st of February Late policy: you will lose 20% of your mark for each day late.
This paper review is not group work. You have to do it on your own.
You can find the following paper on the wiki page with the course notes: Lossless Re-encoding of JPEG Images Using Block-adaptive Intra Prediction (Matsuda et al, 2008)
Read the abstract of the paper. It is very unlikely that you will understand what it's about, or even know the meaning of individual words in the abstract, such as:
- transcoding [scheme]
- block-adaptive intra prediction
- spacial domain
- DCT domain|coefficients
- prediction residuals
- block-based classification
- probability density functions|model
- multisymbol arithmetic coder
- entropy coding
- coding rates
Obviously this was written by (and for) experts in the field of digital image processing and they have years/decades of experience working with all these things. But let's pretend you really needed to understand (to some degree) what problem they were trying to solve and how their solution worked.
You will need to submit a document (probably a couple of pages) with:
- A rewritten abstract using language that someone like you before you did this assignment could understand. Do not drop significant information. This will likely end up beeing longer than the original abstract.
- For 8 of the terms listed above write a description, about a paragraph long, explaining what it means. Also mention how you learned it yourself (what resources such as websites you used, please put the URLs of the websites at the bottom of the paper). If you're having a hard time filling a paragraph with the explanation - consider adding an example, or comparison to a simpler technique/concept. Obviously you need to quote any text you copy (or slightly modify). Please make sure that you're using your own words as much as possible.
- In the last paragraph please explain what parts of this assignment were especially difficult for you. If you submit your paper before the deadline I will have time to account for your feedback in the next paper review.
You don't need to follow academic standards such as MLA but your paper should look professional, like something you would submit to your boss's boss at work (assuming you're working in an old-school formal organisation).
Save your file as an ODT or DOC or DOCX. I will appreciate it if you make the links clickable. Submit to Moodle before the deadline.
Grading: 10 points: 1 for the rewritten abstract, 1 point for each of the 8 terms explained, and 1 point for the look of the document.