Object recognition is a hot topic in the field
of computer vision that arises the interest of many researchers in the field.
Any object recognition algorithm relies heavily on a good pair of detectors
/descriptors that affect the discrimination power of any classifier used to
recognize those objects and categorize them. For many years the recognition task
was mainly relying on 2D images and therefore was using 2D features.
However, with the appearance of 3D cameras, such as the Kinect,
other descriptors have come into play in order to take advantage of the 3D
information brought by these cameras. Our main research interest is therefore to
work on 3D Detectors/Descriptors for tasks such as object recognition, scene
understanding and scene matching. Also, same ideas can be used for human pose
The basic workflow followed for doing this is first scene segmentation, second
feature detection/description and finally object classification.