Search results for "NETWORKING"

Incorporating depth information into few-shot semantic segmentation

2021

International audience; Few-shot segmentation presents a significant challengefor semantic scene understanding under limited supervision.Namely, this task targets at generalizing the segmentationability of the model to new categories given a few samples.In order to obtain complete scene information, we extend theRGB-centric methods to take advantage of complementary depthinformation. In this paper, we propose a two-stream deep neuralnetwork based on metric learning. Our method, known as RDNet,learns class-specific prototype representations within RGB anddepth embedding spaces, respectively. The learned prototypesprovide effective semantic guidance on the corresponding RGBand depth query ima…

A Mobile Computing Framework for Pervasive Adaptive Platforms

2012

International audience; Ubiquitous computing is now the new computing trend, such systems that interact with their environment require self-adaptability. Bioinspiration is a natural candidate to provide the capability to handle complex and changing scenarios. This paper presents a programming framework dedicated to pervasive platforms programming. This bioinspired and agentoriented framework has been developed within the frame of the PERPLEXUS European project that is intended to provide support for bioinspiration-driven system adaptability. This framework enables the platform to adapt itself to application requirements at high-level while using hardware acceleration at node level. The resu…

Embedded multi-spectral image processing for real-time medical application

2016

International audience; The newly introduced Kubelka-Munk Genetic Algorithm (KMGA) is a promising technique for the assessment of skin lesions from multi-spectral images. Using five skin parameter maps such as concentration or epidermis/dermis thickness, this method combines the Kubelka-Munk Light-Tissue interaction model and Genetic Algorithm optimization process to produce a quantitative measure of cutaneous tissue. Up to the present, variant improved KMGA implementations have been successfully realized using the recent parallel computing techniques. However, all these achievements are based on the multi-core CPUs. This results in a quite high cost and low practicability for the hardware …

Network architectures and energy efficiency for high performance data centers

2017

The increasing trend to migrate applications, computation and storage into more robust systems leads to the emergence of mega data centers hosting tens of thousands of servers. As a result, designing a data center network that interconnects this massive number of servers, and providing efficient and fault-tolerant routing service are becoming an urgent need and a challenge that will be addressed in this thesis. Since this is a hot research topic, many solutions are proposed like adapting new interconnection technologies and new algorithms for data centers. However, many of these solutions generally suffer from performance problems, or can be quite costly. In addition, devoted efforts have n…

Arbres couvrants presque disjoints

2015

International audience; Dans un réseau, la recherche de plusieurs arbres couvrants avec des propriétés intéressantes a amené à l'introduc-tion de plusieurs notions : les arbres couvrants arête-disjoints, les arbres indépendants enracinés en un sommet et les arbres complètement indépendants. Afin de généraliser ces notions, nous introduisons la notion d'arbres couvrants (i, j)-disjoints, où i et j sont respectivement le nombre maximum de noeuds internes et d'arêtes communs aux arbres couvrants. Nous montrons que déterminer s'il existe deux arbres couvrants (i, j)-disjoints dans un graphe G est un problème NP-complet pour i et j quelconques, et nous déterminons les valeurs minimales de i et j…

Presentations of constrained systems with unconstrained positions

2005

International audience; We give a polynomial-time construction of the set of sequences that satisfy a finite-memory constraint defined by a finite list of forbidden blocks, with a specified set of bit positions unconstrained. Such a construction can be used to build modulation/error-correction codes (ECC codes) like the ones defined by the Immink-Wijngaarden scheme in which certain bit positions are reserved for ECC parity. We give a lineartime construction of a finite-state presentation of a constrained system defined by a periodic list of forbidden blocks. These systems, called periodic-finite-type systems, were introduced by Moision and Siegel. Finally, we present a linear-time algorithm for con…

A Modularity Backbone Extraction Method for Weighted Complex Networks

2022

The exponential growth in the size of real-world networks is a major barrier to analyzing their structure and dynamics. Thus, reducing the network's size while maintaining its topological features is highly significant. As community structure is one of the fundamental fingerprints of real-world networks, this work proposes a new node-filtering backbone extraction method to preserve the network's community structure.

QBAIoT: QoS Based Access for IoT Environments

2018

Autonomic Computing and VANETs

2017

Robust control of networks under discrete disturbances and controls

2011

We consider dynamic networks where the disturbances and control actions take discrete values. We briefly survey some of our recent results establishing necessary and sufficient conditions for the existence of robustly globally invariant (hyper box) sets, as well as sufficient conditions for global attractivity of such sets.We then establish connections between these results and existing results in the literature for the setup where all the inputs are analog. Finally, we derive tight upper and lower bounds on the smallest such set in the special case of a degenerate network.