Abstract. We present a novel shape recognition method based on an algorithm to detect contrasted level lines for extraction, on Shape Context for encoding and on an a contrario app...
This paper compares two methods for object localization from contours: shape context and chamfer matching of templates. In the light of our experiments, we suggest improvements to...
Arasanathan Thayananthan, Bjoern Stenger, Philip H...
Action recognition is one of the most active research fields in computer vision. In this paper, we propose a novel method for classifying human actions in a series of image seque...
We propose a fast algorithm, EMD-L1, for computing the Earth Mover's Distance (EMD) between a pair of histograms. Compared to the original formulation, EMD-L1 has a largely si...
Feature-based methods have found increasing use in many applications such as object recognition, 3D reconstruction and mosaicing. In this paper, we focus on the problem of matchin...