Work ow research at HP Labs has evolved through three stages during the past four years. Our e ort started with the enhancement of an existing HP work ow product { WorkManager. Then, based on customers' feedback, a new work ow prototype was developed to support the requirements of enterprise complex business process management. This formed the base for HPs second generation work ow product { Ad...

We compare two multicommodity ow problems, the maximum sum of ow, and the maximum concurrent ow. We show that, for a given graph and a given set of k commodities with speci ed demands, if the minimum capacity of a multicut is approximated by the maximum sum of ow within a factor of , for any subset of commodities, then the minimum cut ratio is approximated by the maximum concurrent ow within a ...

In this paper we discuss the formulation of the governing equations that describe ow of uids in porous media Various types of uid ow ranging from single phase ow to compositional ow are considered It is shown that all the di erential equations governing these types of ow can be e ectively rewritten in a fractional ow formulation i e in terms of a global pressure and saturation or saturations an...

Abstract The Mallard (Anas platyrhynchos) duck is a ubiquitous and socio-economically important game bird in North America. Despite their generally abundant midcontinent population, Mallards eastern America are declining, which may be partially explained by extensive hybridization with human-released domestically derived game-farm Mallards. We investigated the genetic composition of middle lowe...

We investigate the complexity and approximability of network ow improvement problems. In these problems, one incurs costs for increasing the capacity of an edge, while the goal is to achieve a ow of maximum value through the network. We study several improvement strategies. Furthermore, we investigate the relationship of network ow improvement problems to network xed cost ow problems, where one...

An approach for learning and estimating temporal-ow models from image sequences is proposed. The temporal-ow models are represented as a set of orthogonal temporal-ow bases that are learned using principal component analysis of instantaneous ow measurements. Spatial constraints on the temporal-ow are also developed for modeling the motion of regions in rigid and coordinated motion. The performa...

Classical data ow analysis determines whether a data ow fact may hold or does not hold at some program point. Probabilistic data ow systems compute a range, i.e. a probability, with which a data ow fact will hold at some program point. In this paper we develop a novel, practicable framework for probabilistic data ow problems. In contrast to other approaches, we utilize execution history for cal...

In this abstract, we present complete-lattice data ow analysis frameworks. These are a generalization of semi-lattice based data ow analysis frameworks and are designed to support the description and solution of data ow analysis problems over ow graphs for concurrent programs. We describe the limitations of existing data ow frameworks as applied to concurrent programs; these limitations are add...

| In this paper, we use single and dualpass satellite-based interferometric radar to measure the ow magnitude of a large alpine glacier. Since the radar measurements alone are not su cient to resolve the 3-D ow direction of the glacier, several ow directions assumptions are made, and checked for mutual consistency. The assumptions that the horizontal component of ow is parallel to the medial mo...

In this paper, we propose a new method for disparityassisted stereo optical ow estimation. This method is based on the linearization of the round-about compatibility constraint, which converts a stereo optical ow estimation problem to a single channel optical ow estimation problem. An over-determined system of optical ow equations can then be constructed for estimating the ow elds. Total least ...