@Inproceedings{WangMansouriCreputRuichek2015_896,
abstract = {We propose the concept of superpixel adaptive segmentation map, to produce a perceptually meaningful representation of rigid pixel image, with higher resolution of more superpixels on interesting regions according to the density distribution of desired attributes. The solution is based on the self-organizing map (SOM) algorithm, for the benefits of SOM's ability to generate a topological map according to a probability distribution and its potential to be a natural massive parallel algorithm. We also propose the concept of parallel cellular matrix which partitions the Euclidean plane defined by input image into an appropriate number of uniform cell units. Each cell is responsible of a certain part of the data and the cluster center network, and carries out massively parallel spiral searches based on the cellular matrix topology. Experimental results from our GPU implementation show that the proposed algorithm can generate adaptive segmentation map where the distribution of superpixels reflects the gradient distribution or the disparity distribution of input image, with respect to scene topology. When the input size augments, the running time increases in a linear way with a very weak increasing coefficient.},
booktitle = {14th Mexican International Conference on Artificial Intelligence (MICAI 2015), Cuernavaca, Morelos, Mexico, Oct. 25-31, 2015, Advances in Artificial Intelligence and Its Applications, Lecture Notes in Computer Science, vol.9414, pp.325-336},
doi = {10.1007/978-3-319-27101-9{\string_}24},
keywords = {Superpixel, Image segmentation, Self-organizing map, Cellular matrix model, Graphics processing unit},
language = {english},
month = oct,
year = 2015,
title = {Massively parallel cellular matrix model for superpixel adaptive segmentation map},
author = {Wang, Hongjian and Mansouri, Abdelkhalek and Creput, Jean-charles and Ruichek, Yassine},
}
@Inproceedings{WangMansouriCreput2015_897,
abstract = {We propose the concept of parallel cellular matrix which partitions the Euclidean plane defined by input data into an appropriate number of uniform cell units. Each cell is responsible of a certain part of the data and the network of the self- organizing map (SOM), and carries out massive parallel spiral searches based on the cellular matrix topology. The advantage of the proposed model is that it is decentralized and based on data decomposition. The required processing units and memory are with linearly increasing relationship to the problem size. Based on the cellular matrix model, the parallel SOM is implemented to deal with various applications including the traveling salesman problem, structured mesh generation, and superpixel adaptive segmentation map. Experimental results of our GPU implementation show that the running time increases in a linear way with a very weak increasing coefficient according to the input size. The proposed cellular matrix model is suitable to deal with large scale problems in a massively parallel way.},
booktitle = {2015 IEEE International Conference on Electronics, Circuits, and Systems (ICECS 2015), Cairo, Egypt, Dec. 06-09},
doi = {10.1109/ICECS.2015.7440384},
keywords = {parallel computation model, self-organizing map, traveling salesman problem, mesh generation, superpixel, GPU implementation},
language = {english},
month = dec,
year = 2015,
title = {Massively parallel cellular matrix model for self-organizing map applications},
author = {Wang, Hongjian and Mansouri, Abdelkhalek and Creput, Jean-charles},
}
@Inproceedings{WangMansouriCreputRuichek2016_929,
abstract = {We propose a distributed local search (DLS) algorithm, which is a parallel formulation of a local search procedure in an attempt to follow the spirit of standard local search metaheuristics. Applications of different operators for solution diversification are possible in a similar way to variable neighborhood search. We formulate a general energy function to be equivalent to elastic image matching problems. A specific example application is stereo matching. Experimental results show that the GPU implementation of DLS seems to be the only method that provides an increasing acceleration factor as the instance size augments, among eight tested energy minimization algorithms.},
booktitle = {International Conference on Swarm Intelligence Based Optimization, ICSIBO'2016, Mulhouse, France, June 13-14. Swarm Intelligence Based Optimization, ICSIBO 2016, Lecture Notes in Computer Science: Volume 10103, pp.65-74, 25 Nov.},
doi = {10.1007/978-3-319-50307-3{\string_}5},
keywords = {Parallel and distributed computing, Variable neighborhood search, Stereo matching, Graphics processing unit},
language = {english},
month = jun,
year = 2016,
title = {Distributed Local Search for Elastic Image Matching},
author = {Wang, Hongjian and Mansouri, Abdelkhalek and Creput, Jean-charles and Ruichek, Yassine},
}
@Article{WangMansouriCreput2017_974,
abstract = {We propose a parallel computation model, called cellular matrix model (CMM), to address large-size Euclidean graph matching problems in the plane. The parallel computation takes place by partitioning the plane into a regular grid of cells, each cell being affected to a single processor. Each processor operates on local data, starting from its cell location and extending its search to the neighborhood cells in a spiral search way. In order to deal with large-size problems, memory size and processor number are fixed as O(N), where N is the problem size. Then one key point is that closest point searching in the plane is performed in O(1) expected time for uniform or bounded distribution, for each processor independently. We define a generic loop that models the parallel projection between graphs and their matching, as executed by the many cells at a given level of computation granularity. To illustrate its efficacy and versatility, we apply the CMM, on GPU platforms, to two problems in image processing: superpixel segmentation and stereo matching energy minimization. Firstly, we propose an extended version of the well-known SLIC superpixel segmentation algorithm, which we call SPASM algorithm, by using a parallel 2D self-organizing map instead of k-means algorithm. Secondly, we investigate the idea of distributed variable neighborhood search, and propose a parallel search heuristic, called distributed local search (DLS), for global energy minimization of stereo matching problem. We evaluate the approach with regards to the state-of-the-art graph cut and belief propagation algorithms. For each problem, we argue that the parallel GPU implementation provides new competitive quality/time trade-offs, with substantial acceleration factors as the problem size increases.},
doi = {https://doi.org/10.1016/j.asoc.2017.08.015},
journal = {Applied Soft Computing},
keywords = {Cellular matrix; Graph matching; k-means; local search; parallel algorithms; graphics processing unit (GPU)},
language = {english},
month = aug,
publisher = {Elsevier},
year = 2017,
title = {Cellular matrix model for parallel combinatorial optimization algorithms in Euclidean plane},
author = {Wang, Hongjian and Mansouri, Abdelkhalek and Creput, Jean-charles},
}
@Inproceedings{MansouriCuiCreputLauri2017_985,
booktitle = {FUTURMOB : pr{\'e}parer la transition vers la mobilit{\'e} autonome, Montb{\'e}liard, France},
keywords = {Parallel and distributed computing, Optical flow, Graphics processing unit},
language = {english},
month = sep,
year = 2017,
title = {Parallelisation of Optical Flow},
author = {Mansouri, Abdelkhalek and Cui, Beibei and Creput, Jean-charles and Lauri, Fabrice},
}
@Inproceedings{CuiMansouriCreput2017_986,
booktitle = {FUTURMOB : pr{\'e}parer la transition vers la mobilit{\'e} autonome, Montb{\'e}liard, France},
language = {english},
month = sep,
year = 2017,
title = {An object tracking parallel algorithm based on multi-frame difference and background subtraction},
author = {Cui, Beibei and Mansouri, Abdelkhalek and Creput, Jean-charles},
}
@Inproceedings{MansouriCreputLauriWang2018_993,
abstract = {optical flow estimation problem of a pair of images using a distributed local search (DLS)},
booktitle = {19{\`e}me congr{\`e}s de la soci{\'e}t{\'e} Fran{\c{c}}aise de Recherche Op{\'e}rationnelle et d'Aide {\`a} la D{\'e}cision, ROADEF 2018},
keywords = {Optical flow, Parallel and distributed computing, Variable neighborhood search, Graphics processing unit.},
language = {english},
month = feb,
year = 2018,
title = {GPU based optical flow estimation using parallel local search algorithm},
author = {Mansouri, Abdelkhalek and Creput, Jean-charles and Lauri, Fabrice and Wang, Hongjian},
}
@Inproceedings{MansouriCreputLauriWang2018_994,
abstract = {The design of many tasks in computer vision field requires addressing difficult NP-hard energy optimization problems. An example of application is the visual correspondence problem of optical flow, which can be formulated as an elastic pattern matching optimization problem. Pixels of a first image have to be matched to pixels in a second image while preserving elastic smoothness constraint on the first image deformation. In this paper, we present a parallel approach to address optical flow problem following the concept of distributed local search. Distributed local search consists in the parallel execution of many standard local search processes operating on a partition of the data. Each process performs local search on its own part of the data such that the overall energy is minimized. The approach is implemented on graphics processing unit (GPU) platform and evaluated on standard Middlebury benchmarks to gauge the substantial acceleration factors that can be achieved in the task of energy minimization.},
award = {/extensions/ICAPManager/pdf/MansouriCreputLauriWang2018_award.pdf},
booktitle = {The Tenth International Conference on Pervasive Patterns and Applications PATTERNS 2018. Barcelona, Spain. February 18-22},
isbn = {978-1-61208-612-5},
issn = {2308-3557},
keywords = {Optical flow, Parallel and distributed computing, Variable neighborhood search, Graphics processing unit.},
language = {english},
month = feb,
note = {Best Paper Award},
year = 2018,
title = {Massively Parallel Optical Flow using Distributed Local Search},
author = {Mansouri, Abdelkhalek and Creput, Jean-charles and Lauri, Fabrice and Wang, Hongjian},
}