We present a novel method for modeling dynamic visual
phenomena, which consists of two key aspects. First, the in-
tegral motion of constituent elements in a dynamic scene is
ca...
Scalable image retrieval systems usually involve hierarchical quantization of local image descriptors, which produces a visual vocabulary for inverted indexing of images. Although ...
Rongrong Ji (Harbin Institute of Technology), Xing...
Abstract. Two approaches have been used to perform approximate inference in Bayesian networks for which exact inference is infeasible: employing an approximation algorithm, or appr...
Recently, stability-based techniques have emerged as a very promising solution to the problem of cluster validation. An inherent drawback of these approaches is the computational c...
In this paper we propose a new framework for modeling 2D shapes. A shape is first described by a sequence of local features (e.g., curvature) of the shape boundary. The resulting ...