Search results for " learning."

showing 10 items of 5179 documents

Learning Flow-Based Feature Warping for Face Frontalization with Illumination Inconsistent Supervision

2020

Despite recent advances in deep learning-based face frontalization methods, photo-realistic and illumination preserving frontal face synthesis is still challenging due to large pose and illumination discrepancy during training. We propose a novel Flow-based Feature Warping Model (FFWM) which can learn to synthesize photo-realistic and illumination preserving frontal images with illumination inconsistent supervision. Specifically, an Illumination Preserving Module (IPM) is proposed to learn illumination preserving image synthesis from illumination inconsistent image pairs. IPM includes two pathways which collaborate to ensure the synthesized frontal images are illumination preserving and wit…

business.industryComputer scienceDeep learningComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONOptical flow02 engineering and technology010501 environmental sciences01 natural sciencesImage (mathematics)Flow (mathematics)Feature (computer vision)Face (geometry)0202 electrical engineering electronic engineering information engineeringCode (cryptography)020201 artificial intelligence & image processingComputer visionArtificial intelligenceImage warpingbusinessComputingMethodologies_COMPUTERGRAPHICS0105 earth and related environmental sciences
researchProduct

Multiscale Attention-Based Prototypical Network For Few-Shot Semantic Segmentation

2021

International audience; Deep learning-based image understanding techniques require a large number of labeled images for training. Few-shot semantic segmentation, on the contrary, aims at generalizing the segmentation ability of the model to new categories given only a few labeled samples. To tackle this problem, we propose a novel prototypical network (MAPnet) with multiscale feature attention. To fully exploit the representative features of target classes, we firstly extract rich contextual information of labeled support images via a multiscale feature enhancement module. The learned prototypes from support features provide further semantic guidance on the query image. Then we adaptively i…

business.industryComputer scienceDeep learningFeature extractionComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]Pattern recognition02 engineering and technologyImage segmentation010501 environmental sciencesSemantics01 natural sciencesImage (mathematics)[INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]Minimum bounding boxFeature (computer vision)0202 electrical engineering electronic engineering information engineering020201 artificial intelligence & image processingSegmentationArtificial intelligencebusiness0105 earth and related environmental sciences
researchProduct

Proposition of Convolutional Neural Network Based System for Skin Cancer Detection

2019

Skin cancer automated diagnosis tools play a vital role in timely screening, helping dermatologists focus on melanoma cases. Best arts on automated melanoma screening use deep learning-based approaches, especially deep convolutional neural networks (CNN) to improve performances. Because of the large number of parameters that could be involved during training in CNN many training samples are needed to avoid overfitting problem. Gabor filtering can efficiently extract spatial information including edges and textures, which may reduce the features extraction burden to CNN. In this paper, we proposed a Gabor Convolutional Network (GCN) model to improve the performance of automated diagnosis of …

business.industryComputer scienceDeep learningFeature extractionPattern recognition02 engineering and technologyFilter (signal processing)OverfittingConvolutional neural network030218 nuclear medicine & medical imaging03 medical and health sciences0302 clinical medicineGabor filter0202 electrical engineering electronic engineering information engineering020201 artificial intelligence & image processingArtificial intelligencebusinessFocus (optics)Spatial analysis2019 15th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)
researchProduct

Combination Of Handcrafted And Deep Learning-Based Features For 3d Mesh Quality Assessment

2020

We propose in this paper a novel objective method to evaluate the perceived visual quality of 3D meshes. The proposed method in no-reference, it relies only on the distorted mesh for the quality estimation. It is based on a pre-trained convolutional neural network (i.e VGG to extract features from the distorted mesh) and handcrafted features extracted directly from the 3D mesh (i.e curvature and dihedral angle). A General Regression Neural Network (GRNN) is used to learn the statistical parameters of the feature vectors and estimate the quality score. Experimental results from for subjective databases (LIRIS masking, LIRIS/EPFL generalpurpose, UWB compression and LEETA simplification) and c…

business.industryComputer scienceDeep learningFeature vectorFeature extraction020207 software engineeringPattern recognition02 engineering and technologyCurvatureConvolutional neural networkVisualizationMetric (mathematics)0202 electrical engineering electronic engineering information engineering020201 artificial intelligence & image processingPolygon meshArtificial intelligencebusiness2020 IEEE International Conference on Image Processing (ICIP)
researchProduct

No-reference mesh visual quality assessment via ensemble of convolutional neural networks and compact multi-linear pooling

2020

Abstract Blind or No reference quality evaluation is a challenging issue since it is done without access to the original content. In this work, we propose a method based on deep learning for the mesh visual quality assessment without reference. For a given 3D model, we first compute its mesh saliency. Then, we extract views from the 3D mesh and the corresponding mesh saliency. After that, the views are split into small patches that are filtered using a saliency threshold. Only the salient patches are selected and used as input data. After that, three pre-trained deep convolutional neural networks are employed for feature learning: VGG, AlexNet, and ResNet. Each network is fine-tuned and pro…

business.industryComputer scienceDeep learningFeature vectorPoolingPattern recognition02 engineering and technology01 natural sciencesConvolutional neural networkResidual neural networkArtificial IntelligenceFeature (computer vision)0103 physical sciencesSignal Processing0202 electrical engineering electronic engineering information engineering020201 artificial intelligence & image processingComputer Vision and Pattern RecognitionArtificial intelligence010306 general physicsbusinessFeature learningSoftwarePattern Recognition
researchProduct

Fallzahlplanung in referenzkontrollierten Diagnosestudien

2002

Purpose: A tutorial illustration of a flexible approach to determine the sample size in reference-controlled diagnostic trials. Materials and Methods: Assuming the usual setting of a new diagnostic method to be compared with a reference method, the emphasis is on the sensitivity of the new method in comparison with the reference method, using a binary outcome (positive versus negative) for both methods. Based on the confidence interval of the sensitivity, a simple but flexible procedure for determining the sample size is described, which incorporates clinically interpretable information. The procedure is illustrated by the fictious planning of a trial to assess the diagnostic value of MRI v…

business.industryComputer scienceDiagnostic TrialMachine learningcomputer.software_genreOutcome (probability)Confidence intervalClinical trialSample size determinationRange (statistics)A priori and a posterioriRadiology Nuclear Medicine and imagingSensitivity (control systems)Artificial intelligencebusinesscomputerRöFo - Fortschritte auf dem Gebiet der Röntgenstrahlen und der bildgebenden Verfahren
researchProduct

Deep Learning for Resource-Limited Devices

2020

In recent years, deep neural networks have revolutionized the development of intelligent systems and applications in many areas. Despite their numerous advantages and potentials, these intelligent models still suffer from several issues. Among them, the fact that they became very complex with millions of parameters. That is, requiring more resources and time, and being unsuitable for small restricted devices. To contribute in this direction, this paper presents (1) some state-of-the-art lightweight architectures that were specifically designed for small-sized devices, and (2) some recent solutions that have been proposed to optimize/compress classical deep neural networks to allow their dep…

business.industryComputer scienceDistributed computingDeep learningIntelligent decision support systemRedundancy (engineering)InitializationArtificial intelligencePruning (decision trees)businessAdaptation (computer science)Quantization (image processing)Convolutional neural networkProceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks
researchProduct

Correlation-Based and Contextual Merit-Based Ensemble Feature Selection

2001

Recent research has proved the benefits of using an ensemble of diverse and accurate base classifiers for classification problems. In this paper the focus is on producing diverse ensembles with the aid of three feature selection heuristics based on two approaches: correlation and contextual merit -based ones. We have developed an algorithm and experimented with it to evaluate and compare the three feature selection heuristics on ten data sets from UCI Repository. On average, simple correlation-based ensemble has the superiority in accuracy. The contextual merit -based heuristics seem to include too many features in the initial ensembles and iterations were most successful with it.

business.industryComputer scienceFeature selectionMachine learningcomputer.software_genreBase (topology)CorrelationComputingMethodologies_PATTERNRECOGNITIONArtificial intelligenceHeuristicsbusinessFocus (optics)Simple correlationcomputer
researchProduct

Improving distance based image retrieval using non-dominated sorting genetic algorithm

2015

Image retrieval is formulated as a multiobjective optimization problem.A multiobjective genetic algorithm is hybridized with distance based search.A parameter balances exploration (genetic search) or exploitation (nearest neighbors).Extensive comparative experimentation illustrate and assess the proposed methodology. Relevance feedback has been adopted as a standard in Content Based Image Retrieval (CBIR). One major difficulty that algorithms have to face is to achieve and adequate balance between the exploitation of already known areas of interest and the exploration of the feature space to find other relevant areas. In this paper, we evaluate different ways to combine two existing relevan…

business.industryComputer scienceFeature vectorSortingRelevance feedbackContext (language use)Machine learningcomputer.software_genreContent-based image retrievalMulti-objective optimizationArtificial IntelligenceSignal ProcessingGenetic algorithmComputer Vision and Pattern RecognitionData miningArtificial intelligencebusinessImage retrievalcomputerSoftwarePattern Recognition Letters
researchProduct

Comprehensive Strategy for Proton Chemical Shift Prediction: Linear Prediction with Nonlinear Corrections

2014

A fast 3D/4D structure-sensitive procedure was developed and assessed for the chemical shift prediction of protons bonded to sp3carbons, which poses the maybe greatest challenge in the NMR spectral parameter prediction. The LPNC (Linear Prediction with Nonlinear Corrections) approach combines three well-established multivariate methods viz. the principal component regression (PCR), the random forest (RF) algorithm, and the k nearest neighbors (kNN) method. The role of RF is to find nonlinear corrections for the PCR predicted shifts, while kNN is used to take full advantage of similar chemical environments. Two basic molecular models were also compared and discussed: in the MC model the desc…

business.industryComputer scienceGeneral Chemical EngineeringMonte Carlo methodLinear predictionGeneral ChemistryLibrary and Information SciencesMachine learningcomputer.software_genreComputer Science ApplicationsRandom forestk-nearest neighbors algorithmMolecular dynamicsNonlinear systemPrincipal component regressionArtificial intelligenceStatistical physicsbusinessConformational isomerismcomputerta116Journal of Chemical Information and Modeling
researchProduct