Today's project by Junian Triajianto is simple on the outside but shows off something that's can be very complex, face detection look easy...
Hi there, this is a new tutorial category in my blog. It's Computer Vision. In this blog, I'd like to show you something cool. It's how to perform Face Detection using your camera / Webcam. You'll see how your application can detect faces from a captured image. Curious about it? Let's take a look how to do that.
This what you need to follow this tutorial:
- Microsoft Visual Studio 2010. Or if you don't have one, you can use 2008 version
- Emgu CV (OpenCV in .NET). You can download the latest version in this link: http://www.emgu.com/wiki/index.php/Main_Page and follow the installation instruction
- Basic Knowledge of C# Programming
- Familiar in WPF Development
After you've got what you need, it's time to rock!
- First thing you should do is installing Emgu CV. Your installation path should be like C:\Emgu\emgucv-windows-x86 22.214.171.1240. And you can see inside C:\Emgu\emgucv-windows-x86 126.96.36.1990\bin some DLLs and sample programs. You can see a simple face detection app Example.FaceDetection.exe and you'll see something like the first picture in this post.
8. Here is the result of our work:
Note : If you can't run your project, just build it and make sure all opencv_xxxx.dll files and haarcascade_frontalface_alt_tree.xml in the same directory with your executable file. You can find those files inside C:\Emgu\emgucv-windows-x86 188.8.131.520\bin .
Take note of his note... Getting the OpenCV and EmguCV dependices were the hardest part of this project.
Also there there's been a file name change to the noted XML file, haarcascade_frontalface_alt_tree.xml. If you get the latest drop of emgucv-windows-x86 the file name's been changed to "haarcascade_frontalface_default.xml". So you'll need to tweak the code to take that into account.
Here's the before;
And the after;
Once the dependences are in place and the file name fixed, this ran like a charm and really did make facial detection look easy...
If you've been wondering how to do face detection, this project is one you'll want to check out...