Many cues have been proposed for contour detection or image segmentation. These include low-level image gradients to high-level information such as the identity of the objects in ...
Model-based recognition of an object typically involves matching dense 3D range data. The computational cost is directly affected by the amount of data of which a transformation n...
We address the problem of object recognition in computer vision. We represent each model and the scene in the form of Attributed Relational Graph. A multiple region representation...
We derive a probabilistic framework for robust, real-time, visual tracking of previously unseen objects from a moving camera. The tracking problem is handled using a bag-of-pixels ...
A fundamental challenge in face recognition lies in determining what facial features are important for the identification of faces. In this paper, a novel face recognition framewo...