Filters








10,976 Hits in 9.3 sec

Real-Time Face Identification via CNN and Boosted Hashing Forest

Yury Vizilter, Vladimir Gorbatsevich, Andrey Vorotnikov, Nikita Kostromov
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
We learn the CNN, then transform CNN to the multiple convolution architecture and finally learn the output hashing transform via new Boosted Hashing Forest (BHF) technique.  ...  Acknowledgement This work is supported by grant from Russian Science Foundation (Project No. 16-11-00082).  ...  Topology Preserving Hashing (TPH) [34] perfroms the Hamming embedding with additional preserving the neighbor ranks.  ... 
doi:10.1109/cvprw.2016.25 dblp:conf/cvpr/VizilterGVK16 fatcat:5wjpjs4rpndn7c3misu5eils7u

AutoCloud+, a "Universal" Physical and Statistical Model-Based 2D Spatial Topology-Preserving Software for Cloud/Cloud–Shadow Detection in Multi-Sensor Single-Date Earth Observation Multi-Spectral Imagery—Part 1: Systematic ESA EO Level 2 Product Generation at the Ground Segment as Broad Context

Andrea Baraldi, Dirk Tiede
2018 ISPRS International Journal of Geo-Information  
, multi-temporal and multi-angular EO big data cubes characterized by the five Vs, namely, volume, variety, veracity, velocity and value.  ...  Never accomplished to date in an operating mode by any EO data provider at the ground segment, systematic ESA EO Level 2 product generation is an inherently ill-posed computer vision (CV) problem (chicken-and-egg  ...  analysis (spatial context-sensitive and spatial topology-preserving) -2D.  ... 
doi:10.3390/ijgi7120457 fatcat:frhng3wbffct5ltnnv6trzofea

AI-WSN: Adaptive and Intelligent Wireless Sensor Network

Gursel Serpen, Jiakai Li, Linqian Liu
2013 Procedia Computer Science  
The characteristics of wireless sensor networks bring many challenges, such as the ultra large number of sensor nodes, dense deployment, changing topology structure, and the most importantly, the limited  ...  As an example, for feed-forward multi-layer neural networks, the training time is mainly dictated by the convergence properties of the specific problem being addressed, which also affects the topology  ...  A wireless sensor network embedded with a neural network can solve a large class problems to develop capability to adapt to changes in a dynamic environment or possess computational intelligence to address  ... 
doi:10.1016/j.procs.2013.09.294 fatcat:hkdcc6q7dvg2vhfqjh5vzyaknm

DeepCloud. The Application of a Data-driven, Generative Model in Design [article]

Ardavan Bidgoli, Pedro Veloso
2019 arXiv   pre-print
Then, we describe the development of a data-driven generative system titled DeepCloud.  ...  It combines an autoencoder architecture for point clouds with a web-based interface and analog input devices to provide an intuitive experience for data-driven generation of design alternatives.  ...  The decoder learns how to reconstruct the original input just by observing this latent representation and, later, can be used to synthesize new output data.  ... 
arXiv:1904.01083v1 fatcat:6qoox2oz75dgrlzprrkhfptaom

An Automated Framework for Multi-label Brain Tumor Segmentation based onKernel Sparse Representation

2017 Acta Polytechnica Hungarica  
The proposed framework is evaluated on the multi-label Brain Tumor Segmentation (BRATS) Benchmark.  ...  Kernel sparse representation, which produces discriminative sparse codes to represent features in a high-dimensional feature space, is the key component of the proposed framework.  ...  This can be solve by the best rank-1 approximation.  ... 
doi:10.12700/aph.14.1.2017.1.3 fatcat:24ps35gxsfeejj66frvxikfrjy

Bridging Algorithm and ESL Design: MATLAB/Simulink Model Transformation and Validation [chapter]

Liyuan Zhang, Michael Glaß, Nils Ballmann, Jürgen Teich
2014 Lecture Notes in Electrical Engineering  
Moreover, commercial tools are available to generate embedded C or HDL code directly from a Simulink model.  ...  We also present a validation technique that considers the functional correctness by comparing the original Simulink model with the generated specification in a co-simulation environment.  ...  Because SysteMoC is originally designed to mainly model data-driven applications, the data dependencies (or topology dependencies in Simulink) are automatically preserved by the FSMs and FIFOs.  ... 
doi:10.1007/978-3-319-06317-1_10 fatcat:zfryj24o7jgsfnwnt7wnmgvr6y

DeepSaliency: Multi-Task Deep Neural Network Model for Salient Object Detection

Xi Li, Liming Zhao, Lina Wei, Ming-Hsuan Yang, Fei Wu, Yueting Zhuang, Haibin Ling, Jingdong Wang
2016 IEEE Transactions on Image Processing  
A key problem in salient object detection is how to effectively model the semantic properties of salient objects in a data-driven manner.  ...  In principle, the proposed saliency model takes a data-driven strategy for encoding the underlying saliency prior information, and then sets up a multi-task learning scheme for exploring the intrinsic  ...  To address these problems, a number of deep learning approaches [22] - [25] have emerged as a powerful tool of data-driven multi-granularity image understanding.  ... 
doi:10.1109/tip.2016.2579306 pmid:27305676 fatcat:6rnohyiqofc2jfeauaq3b2cllq

State of the Art in Surface Reconstruction from Point Clouds [article]

Matthew Berger, Andrea Tagliasacchi, Lee M. Seversky, Pierre Alliez, Joshua A. Levine, Andrei Sharf, Claudio T. Silva
2014 Eurographics State of the Art Reports  
The traditional problem addressed by surface reconstruction is to recover the digital representation of a physical shape that has been scanned, where the scanned data contains a wide variety of defects  ...  This state-of-the-art report surveys the field of surface reconstruction, providing a categorization with respect to priors, data imperfections, and reconstruction output.  ...  Pierre Alliez is supported by an ERC Starting Grant "Robust Geometry Processing" (257474).  ... 
doi:10.2312/egst.20141040 fatcat:vuosd5taxnhcbaronplzvv4oqi

The meshing framework ViennaMesh for finite element applications

Florian Rudolf, Josef Weinbub, Karl Rupp, Siegfried Selberherr
2014 Journal of Computational and Applied Mathematics  
Further contributing to this problem is the lack of a common programming interface, impeding convenient switching of meshing backends.  ...  ViennaMesh tackles this challenge by providing a uniform meshing interface and reusable mesh-related tools, like CGAL, Gmsh, Netgen, and Tetgen.  ...  Acknowledgment This work has been supported by the European Research Council (ERC) through the grant #247056 MOSILSPIN.  ... 
doi:10.1016/j.cam.2014.02.005 fatcat:rqhfgwrm7nf4xo6hayxfmnrif4

BDA-SketRet: Bi-Level Domain Adaptation for Zero-Shot SBIR [article]

Ushasi Chaudhuri, Ruchika Chavan, Biplab Banerjee, Anjan Dutta, Zeynep Akata
2022 arXiv   pre-print
Finally, our CNN-based model confirms the discriminativeness of the shared latent space through a novel topology-preserving semantic projection network.  ...  The efficacy of zero-shot sketch-based image retrieval (ZS-SBIR) models is governed by two challenges.  ...  In this paper, we aim to solve the (G)ZS-SBIR problem by looking into some of the existing problems that are explained in the following subsections and propose improved solutions to this end.  ... 
arXiv:2201.06570v1 fatcat:jnqwikdwjvhy7fo3kvcf6frn4q

Introduction to genetic programming tutorial

John R. Koza
2010 Proceedings of the 12th annual conference comp on Genetic and evolutionary computation - GECCO '10  
G The result solves a problem of indisputable difficulty in its field.  ...  the data 164 TURING'S THREE APPROACHES TO MACHINE INTELLIGENCE • Turing made the connection between searches and the challenge of getting a computer to solve a problem without explicitly programming it  ...  discover and reuse subprograms in the course of automatically creating computer programs to solve problems. 1999 • Genetic programming possesses the attributes that can reasonably be expected of a system  ... 
doi:10.1145/1830761.1830894 dblp:conf/gecco/Koza10 fatcat:zzlqcyt27bekthlg4mdswhkv6e

Fusing Deep Learning and Sparse Coding for SAR ATR

Odysseas Kechagias-Stamatis, Nabil Aouf
2018 IEEE Transactions on Aerospace and Electronic Systems  
We propose a multi-modal and multi-discipline data fusion strategy appropriate for Automatic Target Recognition (ATR) on Synthetic Aperture Radar imagery.  ...  Evaluation on the MSTAR dataset yields the highest ATR performance reported yet which is 99.33% and 99.86% for the 3 and 10-class problems respectively.  ...  IV Experimental results on the MSTAR data set under various configurations such as the 10-class ATR problem with and without target variants, the 3-class ATR problem, and affected by several nuisance factors  ... 
doi:10.1109/taes.2018.2864809 fatcat:lsn5aiopobfvtoohy6ybasooby

Predicting Brain Multigraph Population From a Single Graph Template for Boosting One-Shot Classification [article]

Furkan Pala, Islem Rekik
2022 arXiv   pre-print
Our MultigraphGNet source code is available at https://github.com/basiralab/MultigraphGNet.  ...  A central challenge in training one-shot learning models is the limited representativeness of the available shots of the data space.  ...  However, all scientific contributions made in this project are owned and approved solely by the authors.  ... 
arXiv:2209.06005v1 fatcat:bvcyml57jvftzjzyeae22sdidu

Meta-simulation of large WSN on multi-core computers

Adnan Iqbal, Bernard Pottier
2010 Proceedings of the 2010 Spring Simulation Multiconference on - SpringSim '10  
The tool flow targets an Occam compiler producing efficient multi-threaded binaries. We expect the final flow of this project to enable sensor code production out of the simulated specification.  ...  The WSN design problem is of high complexity, and requires robust methodologies, including simulation support.  ...  Grid computing is also becoming a mature technique and is being used to solve many scientific computing problems.  ... 
doi:10.1145/1878537.1878676 fatcat:penbjrh2nvaethhd5od7m3eu74

Topology-Aware Performance Optimization and Modeling of Adaptive Mesh Refinement Codes for Exascale

Cy P Chan, John D Bachan, Joseph P Kenny, Jeremiah J Wilke, Vincent E Beckner, Ann S Almgren, John B Bell
2016 2016 First International Workshop on Communication Optimizations in HPC (COMHPC)  
Furthermore, we show that network latency in the multigrid bottom solve is the main contributing factor preventing good scaling on exascale-class machines.  ...  Mota Mapper generates multiobjective, network topology-aware box mappings, which we apply to optimize the data layout for the example multigrid solvers.  ...  To understand the problem the monad solves, consider that data-independent expression trees are all static and finite.  ... 
doi:10.1109/comhpc.2016.008 dblp:conf/sc/ChanBKWBAB16 fatcat:g3moavjnpfcmfplyazisu2etaa
« Previous Showing results 1 — 15 out of 10,976 results