Another course with potential, ruined by a flakey operational platform. IBM Cloud as provisioned in this course is not fit for purpose, end of story. Exercises are often impossible to do because servers, storage or proxies in the IBM cloud are down, and the instructions are not always complete either.

Sadly, the material, though informative at a very high level, is nowhere near detailed enough to actually teach the underlying integration, though you'll be well able to cut, paste and press "Shift-Enter" by the time you finish, even if you have no clue what the code you're executing does.

The title is misleading too - this is a Watson Visual Recogniser course. It describes this product (as a visual classifier) quite well, but it barely touches any of the ideas behind Computer Vision, and doesn't really teach anything at all about OpenCV.

Worth doing for free if you have nothing better to do, but there are better ways to spend your time and money.

Too many technical hurdles, too little help, disappointed with the level of support

Except the week 3, this is only advertising for Watson from IBM.

The third weeks is teaching interesting things about OpenCV and the other week are just saying every minutes how Watson is great but the using could be teach in 5 minutes.

This is the first pathetic course experience I have on Course - my assignment has not been reviewed since last 4 days . . .

The target material for this course is very interesting. The execution of the course is severely lacking. This course is in desperate need of quality control review by someone other than the instructors in order to point out everything that needs to be fixed.

Deprecated commands are used in the jupyter notebooks.

The instructions for the final talk about evaluating furniture, when in reality you're supposed to be evaluating helmets. The "supplied images" for helmet/no-helmet do not appear until AFTER the project is submitted, so anyone who used their own images for helmet/no-helmet technically gets failed for not using the supplied images, which are not supplied up-front.

The instructions for publishing the classification model to the web is full of errors as can be witnessed by the desperate conversations in the forum.

I voluntarily peer-reviewed more than the required (single) final assignment and the more that I reviewed, the more I could see that the above problems are being experienced by many with very little useful resolution from the course overlords.

Please... this entire course... have someone go through it from beginning to end and debug it. The material being covered is GREAT, but the execution leaves a lot to be desired. If this had been the first course in the specialization I was taking, I would have given up. It's that bad.

Course was good but not the content. Almost every lab is broken and people have to wait for the course administrator to respond. At the end we ourselves join the things so that it will work so on one hand its nice that we learn as things are broken but for some people its first hand to programming or IBM Watson and the image they are going to carry will not be nice.

For the coursera I would strongly comment that this course is full of errors and how they placed this course with the monthly subscription be the big question. For me if it be free there will be some relaxation.

I set up a 1 star because, in my opinion, it is the worst course that I ever attended:

I already had several major technical issues during week 1 week 3 interval, like:

1. IBM Lab environment doesn't work;

2. Links to images don't work;

3. I can not create an IBM Cloud account because of the error that occurs every time for no reason;

4. The worst thing is that the support team doesn't respond to the discussion forum;

It was a great course that I had fun completing.

I thank the instructors and especially the course mentors for their constant and timely support.

Godspeed,

Tanisha Cijo.

About 90% of this course is lab assignments. The labs are too unstable and non-responsive. Lab instructions needs to be updated. I had to depend on the teaching assistant's responses in the discussion forum for almost all of the labs. As per my observation, majority of the resolution the teaching staff provided too is to use a different browser and clear the history/cache and restart the labs.

Much of the course talks about Watson, though the course title includes OpenCV.

This is one of the worst experiences I have had in any of the online courses I have taken so far.

Too many bugs in the course to work through.

Labs are quite poorly written and make it very difficult to learn.

Very specific to using only Watson; not much math is explained. This course could be great for some audience, but not for me. I need a course that gives me deeper insight.

most of the codes is not explained , late support to solve it .

I was looking forward to taking this class, but I'm very disapointed. I'm only halfway through it and really struggle to continue. The instructor just reads his text like a robot, transitioning from one subject to another in the same sentance and without any change in tone, making it all really confusing. It's the same for the slides: first, they are not great. And also, sometimes a slide only appears for a split second, but since the instructor takes absolutely no pause between two subjects, it leaves no time to even catch a glimpse of its content. The provided notebooks are also very repetitive, contain some errors, and contain either too many details (ex. PIL/openCV in week 2) or too little. And finally, IBM CV Studio's app seems cool, but I would first like to learn how to do CV on my own through code rather than through a proprietary web app.

This course is a perfect starter guide for those interested in Computer Vision. Clear, crisp, and interesting,, this course gives the student all the information they need, re-enforced with examples, and a variety of applications. Neither too hard an complicated, nor to easy and boring, this course is perfectly in the middle, and is an engaging resource for many.

Although the topic was interesting and the application relatively easy, there were too many obstacles to really enjoy the rollout of the classifier in the cloud.

This course had many failures that wasted a great deal of my time. To labs were out of service for over a week, making at least five of them unusable. The video lessons would often throw out terms without explaining what they meant, how they could be used, their advantages or drawbacks. If you are going to bring up a complex subject, you owe your audience some sort of explanation. If not, then don't use the term. I found I had to supplement the lectures with deep research on my own to find out things that could have been explained up front. The reader was lifeless, and brought nothing to the written text of the lectures. This is the worse course I've ever taken on Coursera.

Well Explained about each and every tools and how to train and test the models and deploy into the cloud services I had very good expreience in collabrating into IBM

The course is very buggy and some of the content and code is wrong, particularly in Week 4 (the capstone project is almost impossible to complete). I contacted Coursera to see if they could help, and they suspended my subscription payments, but otherwise said it was not their problem. If you have to take this course, be very aggressive about the schedule so that you have time at the end to deal with all the technical problems. Also, make sure you have a look at the Forums and try to find and/or post any issues you have; Technical Support does their best to help. There are also some "surprises" about how things work in the capstone project that you will only find out about in the Forums. BTW, there are other learning sites that are much better and cheaper.

It was a really poorly planned course in terms of the instructions given and solutions offered when troubleshooting.

Some of the material was outdated where the current UI does not match the instructions given.

In the last task, however, was the worst. After asking three times for a solution to an issue that multiple students faced, there was no response. There is still no solution after the initial question was asked by another student two weeks ago.

Took as part of IMB AI Engineering Preoffresional Certficate. Worst course in the program so far. Materials seemed confused without consitent style/tools. Some of the projects and materials using the Compuer Vision Studio and IMB cloud services didnt work. You basiclally just end up cutting and pasting code in various web tools.

Total waste of time.

The lessons are informational, the quizzes are aligned to check a rough level of basic understanding.

However the labs have barely any relationship, they are far away more advanced than the lesson: they have no learning purpose, there's nothing to experiment.

The code using open-cv and keras should eb straightforward to illustrate the lessons and experiment: both of these libraries work pretty well. However the purpose of this class is to showcase IBM products wrapped on the top of these free libraries, hence adding a level of complexity which should not exist.

Besides, some code is not working, showing that the labs were not reviewed by the staff.

I stopped at the end of the last week when it was time to "deploy" my app thru the IBM product. I spent zero time working on open-cv or on a CNN or even any other ML... I spent time fighting with the wrapping code and IBM app.

My take away based on the title of the course: "introduction to computer vision and image processing".. yes, that was an introduction to talk about it at the office while sipping a coffee. That's pretty much about it.

Not enough focus on Image Detection, Classification and Localization. Many models are already trained and you just evaluate the results. At least, you get to know the names of models that are used and how to execute the code.

The CV Studio platform is broken (which is mandatory to use) and does NOT allow to deploy the model into the Final Exercise Submission. This was VERY sketchy!

Thankfully, the Discussion Forum and Coursera Forum Moderators are very helpful!

Support by IBM on the CV Studio platform is non-existent, the do not even reply, some Lab exercises (2 or 3) are broken and neglected.

Worst course in the learning path so far

Learning content was good, but making assignments was a nigthmare, I could not set up an IBM Cloud account, becasue they could not verify my payment card, when i asked them to help me, they just responded "We have reviewed your account/transaction and will not be able to offer services. No further information will be disclosed regarding this matter. Any card authorizations will reverse within 24-72 hours depending on the issuing bank.". They ghosted me completely and did not want to help. I do not recommend. It's no wonder that with this kind of customer attitude IBM is far behind the competition today.

Compared to the other two courses I took this had very poor support. None of the labs worked as the course author/instructor claimed it works. Course material is difficult and there is no one who can answer questions and tell us why the labs are not working the way they should. Unless coursera improves this course I would not recommend any one take this course. You will have to learn on your own.