We consider simulations of graph automata. We introduce two local transformations on the neighborhood: splitting and merging. We explain how to use such transformations, and their consequences on the topology of the simulated graph, the speed of the simulation and the memory size of simulating automata in some cases. As an example , we apply these transformations to graph automata embedded on s...

This paper proposes novel metrics to evaluate the performance of object detection algorithms in video sequences. The proposed metrics allow to characterize the methods being used and classify the types of errors into region splitting, merging or merge-split, detection failures and false alarms. This methodology is applied to characterize the performance of five segmentation algorithms. These te...

We give an axiomatization of the Aumann-Shapley cost-sharing method in the discrete case by means of monotonicity and no merging or splitting (Sprumont, 2005 [16]). Monotonicity has not yet been employed to characterize this method in such a case, unlike the case in which goods are perfectly divisible, for which Monderer and Neyman (1988) [10] and Young (1985) [19] characterize the Aumann-Shapl...

This paper presents a performance metric for the document structure extraction algorithms by finding the correspondences between detected entities and ground truth. We describe a method for determining an algorithm’s optimal tuning parameters. We evaluate a group of document layout analysis algorithms on 1600 images from the UW-III Document Image Database, and the quantitative performance measu...

Automatizing the merging of knowledgebases is an important step towards more efficient knowledge management. The cases when two knowledgebases need to be merged completely into a monolithic result, however, are relatively rare. Most often, some of the information is irrelevant, not trusted, or needs special treatment as a belief, opinion or preference. This paper presents an approach for partia...

Planar samples of n-GaAs with attached point contacts at different dc voltages may display a variety of spatiotemporal patterns arising from the dynamics of curved charge dipole waves. Patterns rank from oscillations due to recycling and motion of simple quasiplanar or cylindrical wave fronts to more complex patterns that include merging and splitting of different fronts. Results of numerical s...

The EM method is used to track moving objects as clusters of pixels significantly different from the corresponding pixels in a reference image. Non-parametric grey-level statistics for the moving clusters are learned online to improve robustness. Simple statistical principles are used for the initialisation, termination, merging and splitting of clusters. The system is complemented by some high...

We show near optimal bounds on the worst case quantum communication of single-shot entanglement-assisted one-way quantum communication protocols for the quantum state redistribution task and for the sub-tasks quantum state splitting and quantum state merging. Our bounds are tighter than previously known best bounds for the latter two sub-tasks. A key technical tool that we use is a convex-split...

Most software agent lifecycles are designed in such a way that they mimic the lifecycles of humans (anthropomorphism). This position paper presents a set of operations for software agents which have no immediate biological analogies, such as splitting and merging agents. We argue through a series of scenarios the potential benefits of this approach. We present a method to implement these operat...

Walter’s comments: Long Range Dependence (LRD) in Internet flows has been known for a long time. Is LRMI just another manifestation of this LRD? LRMI looks at information inside a packet instead of just the “aggregate” statistics like “flow sizes” but what more information does it convey beyond LRD. Since LRMI is (or can be) defined on a per-flow basis, it might convey more information about th...