Saturday, October 26, 2013

OpenCV Computer Vision Application Programming: Review

I was pleasantly surprised when I heard that there is a video tutorial for learning OpenCV out in the market.

Its available for download at http://bit.ly/1eJjrWN. A preview of the video can be seen at http://bit.ly/1gpqpPW.

I did see a few videos on YouTube for "OpenCV Tutorials", but they weren't organized properly. Its good to see an author walk you through all the basic modules of OpenCV. The author discusses about OpenCV installation, basic C, C++ and Python interfaces. There is a video on Face Recognition, Image Morphology, Image segmentation, Calibration, Training your own HAAR cascade classifier, etc. I specifically looked into "Working with HDR Images" video and was disappointed. The author only talked about how to use a ready made program in Linux and how to save HDR images using the built program. It is sad that OpenCV is already more updated after the release of this video (like saving HDR images programatically http://docs.opencv.org/trunk/doc/tutorials/photo/table_of_content_photo/table_of_content_photo.html).

The instructions in the video focus primarily on using the OpenCV interface. If you want to know why are do you want to use a certain instruction/statement, you will have learn Image Processing, Machine Learning/Pattern Recognition. The code written by the author is available for download along with the videos.

Although, I would have liked the author to place all the code in a single folder and explain using that organized folder (I could see the author only opening up random files, executing binary, explaining output and explaining the code.

The programming has been primarily done in C++ which is nice. Most of developers prefer C++ for their development. Its also easier to port the code to mobile or across different platforms. It is a good addition to Packt's OpenCV book http://www.packtpub.com/opencv-2-computer-vision-application-programming-cookbook/book .

P.S. I received a free copy of video tutorial of "OpenCV Computer Vision Application Programming" for writing the review.

No comments:

Post a Comment