Search results for " image processing."

showing 10 items of 2265 documents

Open Set Audio Classification Using Autoencoders Trained on Few Data.

2020

Open-set recognition (OSR) is a challenging machine learning problem that appears when classifiers are faced with test instances from classes not seen during training. It can be summarized as the problem of correctly identifying instances from a known class (seen during training) while rejecting any unknown or unwanted samples (those belonging to unseen classes). Another problem arising in practical scenarios is few-shot learning (FSL), which appears when there is no availability of a large number of positive samples for training a recognition system. Taking these two limitations into account, a new dataset for OSR and FSL for audio data was recently released to promote research on solution…

Computer scienceOpen set02 engineering and technologylcsh:Chemical technologyMachine learningcomputer.software_genreBiochemistryArticleAnalytical ChemistrySet (abstract data type)open set recognition020204 information systemsaudio classificationautoencoders0202 electrical engineering electronic engineering information engineeringFeature (machine learning)lcsh:TP1-1185few-shot learningElectrical and Electronic EngineeringRepresentation (mathematics)Instrumentationbusiness.industryopen set classificationPerceptronClass (biology)AutoencoderAtomic and Molecular Physics and OpticsEmbedding020201 artificial intelligence & image processingArtificial intelligenceTransfer of learningbusinesscomputerSensors (Basel, Switzerland)
researchProduct

Full-parallax 3D display from stereo-hybrid 3D camera system

2018

Abstract In this paper, we propose an innovative approach for the production of the microimages ready to display onto an integral-imaging monitor. Our main contribution is using a stereo-hybrid 3D camera system, which is used for picking up a 3D data pair and composing a denser point cloud. However, there is an intrinsic difficulty in the fact that hybrid sensors have dissimilarities and therefore should be equalized. Handled data facilitate to generating an integral image after projecting computationally the information through a virtual pinhole array. We illustrate this procedure with some imaging experiments that provide microimages with enhanced quality. After projection of such microim…

Computer sciencePoint cloudComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION02 engineering and technologyStereo display01 natural sciences010309 opticsComputer graphics (images)0103 physical sciences0202 electrical engineering electronic engineering information engineeringComputer visionElectrical and Electronic EngineeringProjection (set theory)Integral imagingbusiness.industryMechanical EngineeringÒpticaViewing angleAtomic and Molecular Physics and OpticsElectronic Optical and Magnetic Materials3d camera020201 artificial intelligence & image processingPinhole (optics)Artificial intelligenceParallaxbusinessOptics and Lasers in Engineering
researchProduct

Dynamic 3D Scene Reconstruction and Enhancement

2017

International audience; In this paper, we present a 3D reconstruction and enhancement approach for high quality dynamic city scene reconstructions. We first detect and segment the moving objects using 3D Motion Segmenta-tion approach by exploiting the feature trajectories' behaviours. Getting the segmentations of both the dynamic scene parts and the static scene parts, we propose an efficient point cloud registration approach which takes the advantages of 3-point RANSAC and Iterative Closest Points algorithms to produce precise point cloud alignment. Furthermore, we proposed a point cloud smoothing and texture mapping framework to enhance the results of reconstructions for both the static a…

Computer sciencePoint cloudComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION02 engineering and technology[ SPI.SIGNAL ] Engineering Sciences [physics]/Signal and Image processingRANSACPoint Cloud Registration0202 electrical engineering electronic engineering information engineeringSegmentationComputer vision3D Scene Enhancement[SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processingComputingMethodologies_COMPUTERGRAPHICSMotion Segmentationbusiness.industry3D reconstruction020207 software engineeringFeature (computer vision)Computer Science::Computer Vision and Pattern Recognition020201 artificial intelligence & image processingArtificial intelligencebusiness3D Reconstruction[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processingTexture mappingSmoothing
researchProduct

CovSel

2018

Ensemble methods combine the predictions of a set of models to reach a better prediction quality compared to a single model's prediction. The ensemble process consists of three steps: 1) the generation phase where the models are created, 2) the selection phase where a set of possible ensembles is composed and one is selected by a selection method, 3) the fusion phase where the individual models' predictions of the selected ensemble are combined to an ensemble's estimate. This paper proposes CovSel, a selection approach for regression problems that ranks ensembles based on the coverage of adequately estimated training points and selects the ensemble with the highest coverage to be used in th…

Computer scienceProcess (computing)Phase (waves)Genetic programming02 engineering and technology01 natural sciencesEnsemble learningSet (abstract data type)010104 statistics & probability0202 electrical engineering electronic engineering information engineering020201 artificial intelligence & image processingPoint (geometry)0101 mathematicsSymbolic regressionAlgorithmSelection (genetic algorithm)Proceedings of the Genetic and Evolutionary Computation Conference
researchProduct

Set similarity joins on mapreduce

2018

Set similarity joins, which compute pairs of similar sets, constitute an important operator primitive in a variety of applications, including applications that must process large amounts of data. To handle these data volumes, several distributed set similarity join algorithms have been proposed. Unfortunately, little is known about the relative performance, strengths and weaknesses of these techniques. Previous comparisons are limited to a small subset of relevant algorithms, and the large differences in the various test setups make it hard to draw overall conclusions. In this paper we survey ten recent, distributed set similarity join algorithms, all based on the MapReduce paradigm. We emp…

Computer scienceProcess (engineering)General EngineeringJoinsScale (descriptive set theory)02 engineering and technologycomputer.software_genreSet (abstract data type)Range (mathematics)Operator (computer programming)Similarity (network science)020204 information systems0202 electrical engineering electronic engineering information engineeringJoin (sigma algebra)020201 artificial intelligence & image processingData miningcomputerProceedings of the VLDB Endowment
researchProduct

Object-Oriented Operational Semantics

2016

Operational semantics is one way of providing meaning to an executable language. On a high level of abstraction, operational semantics means to define an interpreter or an abstract machine for the language. In this article, we review the concept of operational semantics in the scope of meta-model-based language definitions and identify challenges and issues. We provide a clean conceptual approach using an object-oriented runtime environment and state change operations, which relies on an underlying abstract virtual machine. We present the approach using a sample language.

Computer scienceProgramming language0102 computer and information sciences02 engineering and technologycomputer.file_formatcomputer.software_genre01 natural sciencesOperational semanticsAbstract machineAction semanticsDenotational semantics010201 computation theory & mathematicsVirtual machine0202 electrical engineering electronic engineering information engineering020201 artificial intelligence & image processingExecutablecomputerInterpreterAbstraction (linguistics)
researchProduct

Car style-holon recognition in computer-aided design

2019

Abstract Multi-scale design can presumably stimulate greater intelligence in computer-aided design (CAD). Using the style-holon concept, this paper proposes a computational approach to address multi-scale style recognition for automobiles. A style-holon is both a whole—it contains sub-styles of which it is composed—as well as a part of a broader style. In this paper, we first apply a variable precision rough set-based approach to car evaluation and ranking. Secondly, we extracted and subsequently computed the each car's characteristic lines from the CAD models. Finally, we identified style-holons using the property of a double-headed style-holon. A style-holon is necessarily included in a t…

Computer scienceProperty (programming)[SHS.INFO]Humanities and Social Sciences/Library and information sciencesComputational MechanicsCAD02 engineering and technologycomputer.software_genre[SHS]Humanities and Social SciencesSet (abstract data type)0203 mechanical engineeringlcsh:TA1740202 electrical engineering electronic engineering information engineeringComputer Aided DesignEngineering (miscellaneous)ComputingMilieux_MISCELLANEOUSbusiness.industryDesign specificationlcsh:Engineering designComputer Graphics and Computer-Aided DesignHuman-Computer InteractionComputational Mathematics020303 mechanical engineering & transportsRankingModeling and Simulation020201 artificial intelligence & image processingArtificial intelligenceRough setHolarchybusinesscomputer
researchProduct

A blind mesh visual quality assessment method based on convolutional neural network

2018

International audience

Computer scienceQuality assessmentbusiness.industry[INFO.INFO-TS] Computer Science [cs]/Signal and Image ProcessingPattern recognition02 engineering and technology01 natural sciencesConvolutional neural network[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing0103 physical sciences0202 electrical engineering electronic engineering information engineering020201 artificial intelligence & image processingArtificial intelligence010306 general physicsbusinessComputingMilieux_MISCELLANEOUS
researchProduct

Computational Offloading in Mobile Edge with Comprehensive and Energy Efficient Cost Function: A Deep Learning Approach

2021

In mobile edge computing (MEC), partial computational offloading can be intelligently investigated to reduce the energy consumption and service delay of user equipment (UE) by dividing a single task into different components. Some of the components execute locally on the UE while the remaining are offloaded to a mobile edge server (MES). In this paper, we investigate the partial offloading technique in MEC using a supervised deep learning approach. The proposed technique, comprehensive and energy efficient deep learning-based offloading technique (CEDOT), intelligently selects the partial offloading policy and also the size of each component of a task to reduce the service delay and energy …

Computer scienceReal-time computingTP1-118502 engineering and technologyBiochemistryVDP::Teknologi: 500::Elektrotekniske fag: 540ArticleAnalytical Chemistry0202 electrical engineering electronic engineering information engineeringcomputational offloadingElectrical and Electronic EngineeringInstrumentationenergy efficiencyMobile edge computingArtificial neural networkbusiness.industryChemical technologyDeep learningdeep learning020206 networking & telecommunicationsEnergy consumptionAtomic and Molecular Physics and OpticsTask (computing)cost functionUser equipment020201 artificial intelligence & image processingmobile edge computingArtificial intelligenceEnhanced Data Rates for GSM Evolutionremote executionbusinessEfficient energy useSensors
researchProduct

Joint Usage of Dynamic Sensitivity Control and Time Division Multiple Access in Dense 802.11ax Networks

2016

It is well known that in case of high density deployments, Wi-Fi networks suffer from serious performance impairments due to hid- den and exposed nodes. The problem is explicitly considered by the IEEE 802.11ax developers in order to improve spectrum efficiency. In this pa- per, we propose and evaluate the joint usage of dynamic sensitivity con- trol (DSC) and time division multiple access (TDMA) for improving the spectrum allocation among overlapping 802.11ax BSSs. To validate the solution, apart from simulation, we used a testbed based on the Wireless MAC Processor (WMP), a prototype of a programmable wireless card.

Computer scienceReal-time computingTime division multiple access050801 communication & media studies02 engineering and technologyFrequency allocation0508 media and communications0202 electrical engineering electronic engineering information engineeringWirelessDense deploymentIEEE 802.11axHidden node problembusiness.industryExposed node problemSettore ING-INF/03 - TelecomunicazioniDynamic sensitivity control05 social sciencesTestbedSpectral efficiencyExposed node problemIEEE 802.11axTDMA020201 artificial intelligence & image processingHidden node problembusinessComputer network
researchProduct