2 comments

[ 3.4 ms ] story [ 14.1 ms ] thread
Author here. Not the best writing on the web, but I'm hoping to spark some discussion. What are some good ways to leverage your own particular aptitude while coming to terms with your shortcomings? How do you determine whether to tackle a problem or accept that it's not worth the time or effort?
I'd recommend you read some of the tutorials here (even if it looks old-school), as it gives visual examples of each algorithm's output: http://homepages.inf.ed.ac.uk/rbf/HIPR2/wksheets.htm

You could also try to read the documentation/source code of some open source image analysis software, and prototype your application around them: http://opencv.org/ http://hugin.sourceforge.net/

Once your prototype works as expected, you can learn more about the algorithms you used, and implement it yourself in your language of choice.

BTW when reading the title of the article I expected something about Debian's package manager, aptitude :)

Edit: to answer to the more general question in the post: if a problem is interesting enough I would invest some time in learning how to solve it.

If it turns out that the domain (image analysis in your case) is not as interesting as I thought, or would require a significant amount of time to learn properly then I'd probably avoid that domain for a while. But I wouldn't give up on it just because I didn't learn about it in school.