Beyond CBIR

In last post, I have written about the series of article about my odyssey in creating the CBIR application in .NET. CBIR is part of Computer Vision. As I was researching in this area, I found Computer Vision has lots of application, rather than just detecting similar images.

The libraries (EMGU CV and Accord Framework) which I introduce during series of post on CBIR, has lot to offer apart from similar image detection.

There are salient articles on the Web if you google for capabilities of both, what I try here is just to give brief  capabilities of both the framework, where you can get the idea and Google more for the application for it.

First in the list is

Face Detection

In my last series of post, I describe about finding the similar images, but those similar images algorithm won’t work for the face detection. Face detection is separate area, which has it’s own set of algorithm, which are available in both Emgu CV and Accord Framework.

Emgu CV Face detection – There is excellent article on the CodeProject, which describe how to detect face using Emgu CV. You can get the sample application of face detection at [Path where EMGU CV is installed]\Emgu.CV.Example\FaceDetection directory.

image Face_Recognition/Main1.jpg

Accord Framework has two application for face detection and tracking. Face detection is for recognizing the face, while tracking is more for gesture control.


Gesture Control

As you have seen in accord framework, it is using face tracking for gesture, both Emgu CV and accord have capability for tracking with face, hand, etc. Here below are youtube videos which does tracking with Emgu CV

Even Accord framework has gesture control, something similar to touch screen device, where you draw the gesture and it identifies, what the gesture is

Object detection

In Emgu CV, there are list of object detection, such as license plate. There is another example, using Accord/Aforge framework to get the traffic signs


Optical character recognition, which every ones familiar, it event been use in the license plate detection. In Accord framework, there is an example of learning deep neural networks using restricted boltzmann machines

Motion Detection

Emgu CV has certain motion detection algorithms, which can get pedestrian from the frame


Photograph Stitching

Emgu CV as well as Accord framework as photo stiching capability. In EmguCV you can get sample application at [Path where EMGU CV is installed]\Emgu.CV.Example\Stitching. For accord framework, sample code is at [Directory where Accord.NET is installed]\Framework\Samples\Imaging\Panorama


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s