—We propose an approach for improving object recognition and localization using spatial kernels together with instance embedding. Our approach treats each image as a bag of insta...
In dynamic spatio-temporal environments where objects may continuously move in space, maintaining consistent information about the location of objects and processing motion-speciï¬...
We frame the problem of object recognition from edge cues in terms of determining whether individual edge pixels belong to the target object or to clutter, based on the configurat...
We investigate a question initiated in the work of Sibley and Wagon, who proved that 3 colors suffice to color any collection of 2D parallelograms glued edge-to-edge. Their proof ...
Many state-of-the-art object recognition systems rely on identifying the location of objects in images, in order to better learn its visual attributes. In this paper, we propose fo...