Tag: computer vision

MSc Thesis of Bence

Oh by the way, I’ve just realised that I forgot to publish my Master’s Thesis here. It’s getting a bit old, pages turning yellow but I can still find typos in it! 🙂


Screenshot 2015-04-12 23.53.06

(Let me know if the download is buggy.)


Online course recommendation

I’m a big fan of online courses, especially the ones that feature good professors with valuable professional and teaching experience. It feels like you are at the university again, but from the comfort (or discomfort) of your chair at home.



I have recently completed “Image and video processing: From Mars to Hollywood with a stop at the hospital” by Guillermo Sapiro.


Image segmentation nodes for ROS

Hello guys!

Check out this part of my (Bence Magyar) MSc thesis work where I created ROS nodes to mask regions in an image based on different attributes.

The nodes can be applied on each other’s output in a pipeline-style way.

They can help to speed up your Computer Vision algorithms by masking the not interesting (defined by you) regions of the images.


Happy segmenting!

The OpenCV, the Qt Creator and the Windows 7

(For the ones who did not get the title: Il buono, il brutto, il cattivo)
After a few hours of googling and hacking we managed to set up OpenCV 2.3 in Qt Creator under Windows 7.

There are a lot of forum and blog posts out there about this but none of them uses the new OpenCV 2.3.

If you are full of reading linking errors like

undefined reference to `cvLoadImage’

you might have just arrived at the right post in the Internet.

After you have downloaded the binaries for Windows, you get a nice README suggesting this:

If you have problems add build\bin and one of build\{x86|x64}\{vc9\vc10\mingw}\bin to your system path (to use DLLs)