Course talk:CPSC522/Conditional GANs for Image to Image Translation

I think this page was generally well-written and was interesting with enough detail and a logical structure, well done! It was good to see examples. I would have liked more on the applications of cGANs maybe in the introduction part? To really motivate solving the image-to-image translations - like what kind of real-world problems are actually getting solved by using images this way? 19/20

(5) The topic is relevant for the course.

(5) The writing is clear and the English is good.

(5) The page is written at an appropriate level for CPSC 522 students (where the students have diverse backgrounds).

(5) The formalism (definitions, mathematics) was well chosen to make the page easier to understand.

(5) The abstract is a concise and clear summary.

(5) There were appropriate (original) examples that helped make the topic clear.

(-) There was appropriate use of (pseudo-) code.

(5) It had a good coverage of representations, semantics, inference and learning (as appropriate for the topic).

(5) It is correct.

(4) It was neither too short nor too long for the topic. (maybe needs more info on motivation? and more explanation in simple language on what the conditioning provides)

(5) It was an appropriate unit for a page (it shouldn't be split into different topics or merged with another page).

(5) It links to appropriate other pages in the wiki.

(5) The references and links to external pages are well chosen.

(5) I would recommend this page to someone who wanted to find out about the topic.

(4) This page should be highlighted as an exemplary page for others to emulate.

I really enjoyed reading about this topic!

The examples are quite nice to show its capabilities,

It would make it a bit clearer if you wrote the variables in latex when describing the objective function.

I would like to perhaps know a bit more about the advantages and disadvantages of this algorithm, I found the last few paragraphs a bit confusing in their structure.

(5) The topic is relevant for the course.

(4) The writing is clear and the English is good.

(5) The page is written at an appropriate level for CPSC 522 students (where the students have diverse backgrounds).

(4) The formalism (definitions, mathematics) was well chosen to make the page easier to understand.

(5) The abstract is a concise and clear summary.

(5) There were appropriate (original) examples that helped make the topic clear.

(-) There was appropriate use of (pseudo-) code.

(4) It had a good coverage of representations, semantics, inference and learning (as appropriate for the topic).

(5) It is correct.

(5) It was neither too short nor too long for the topic.

(5) It was an appropriate unit for a page (it shouldn't be split into different topics or merged with another page).

(5) It links to appropriate other pages in the wiki.

(5) The references and links to external pages are well chosen.

(5) I would recommend this page to someone who wanted to find out about the topic.

(4) This page should be highlighted as an exemplary page for others to emulate.

In the abstract, tell us what "performed in a conditional setting" means.

I don't get the paragraph in the GANs section on loss (the last paragraph in the version I am looking it). You should try to explain why a loss is not needed. (Why isn't the "Is D correct" in the figure providing a loss function?) Later you say (for CGANS) "also to minimize the loss between the generated image and the expected target image" - isn't this the same loss you said doesn't exist? Then you give "loss function for a conventional GAN", so I'm even more confused.

You need to provide an explicit reference for the figures that are not your's on the main page. Ypu don't want anyone to think you are claiming something is your's, when it isn't. Eg. the "High-level structure of a GAN. " figure is not your's.

"However, GANs generate random images in the output domain." is the word "random" used in a informal meaning, or is it saying something about the distribution of images in the output domain? In cases where a statenent can be misunderstood, try to be more accurate in what you mean. (Which is why we ask for feedback; it tells us what can be misunderstood.)

"learns a mapping from x to z" I think should be "learns a mapping from x to y". What is preventing it from still doing that?

Try to give a bigger picture in the Discussion. Are cGANS only useful for image-to-image translation tasks? What else could they be used for? Why would anyone who is not interested in image-to-image translation tasks be interested in them?