Correlation is a standard technique for recognizing known patterns in two-dimensional grid (pixel) images. Its obvious importance has led to numerous hardware implementations and variations. Images captured directly onto 3D voxel grids are becoming more common, including those from confocal microscopy and medical imaging technologies. To our knowledge, no one has yet addressed correlation as a technique for recognizing 3D templates in such 3D voxel data. We find that this problem includes a number of issues: efficient three-axis rotation of a template with respect to 3D image, large volume of results from the correlation, and the possibility of a template matching an image multiple times. We briefly review techniques that have been used in 2D template matching, and examine analogies to a molecule interaction problem in computational chemistry, including its similarity to multispectral images. We report on a hardware accelerator for the 3D correlation problem, based on a commodity coprocessor board containing field programmable logic arrays (FPGAs). Because the convolution processor is built from reconfigurable logic, it can be adapted to non-linear scoring algorithms using complex data values at each voxel, and can be tailored to solve other problems such as anisotropic grid axes. We present initial performance results for the FPGA implementation, and note that accelerator performance is likely to grow roughly linearly with FPGA capacity, process improvements, and number of FPGAs.
{"title":"Three-dimensional template correlation: object recognition in 3D voxel data","authors":"T. Court, Y. Gu, M. Herbordt","doi":"10.1109/CAMP.2005.52","DOIUrl":"https://doi.org/10.1109/CAMP.2005.52","url":null,"abstract":"Correlation is a standard technique for recognizing known patterns in two-dimensional grid (pixel) images. Its obvious importance has led to numerous hardware implementations and variations. Images captured directly onto 3D voxel grids are becoming more common, including those from confocal microscopy and medical imaging technologies. To our knowledge, no one has yet addressed correlation as a technique for recognizing 3D templates in such 3D voxel data. We find that this problem includes a number of issues: efficient three-axis rotation of a template with respect to 3D image, large volume of results from the correlation, and the possibility of a template matching an image multiple times. We briefly review techniques that have been used in 2D template matching, and examine analogies to a molecule interaction problem in computational chemistry, including its similarity to multispectral images. We report on a hardware accelerator for the 3D correlation problem, based on a commodity coprocessor board containing field programmable logic arrays (FPGAs). Because the convolution processor is built from reconfigurable logic, it can be adapted to non-linear scoring algorithms using complex data values at each voxel, and can be tailored to solve other problems such as anisotropic grid axes. We present initial performance results for the FPGA implementation, and note that accelerator performance is likely to grow roughly linearly with FPGA capacity, process improvements, and number of FPGAs.","PeriodicalId":393875,"journal":{"name":"Seventh International Workshop on Computer Architecture for Machine Perception (CAMP'05)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134001931","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Giuseppe Boccignone, V. Caggiano, A. Marcelli, Paolo Napoletano, G. D. Fiore
We present an architectural model for the interaction between top-down, object-based information and bottom-up, spatial-based information in determining visual attention shifts. We focus in particular on how the attentive process can take into account the processing of faces and multiple moving objects. To validate the model, experiments with eye-tracked human subjects are presented and discussed.
{"title":"An architectural model for combining spatial-based and object-based information for attentive video analysis","authors":"Giuseppe Boccignone, V. Caggiano, A. Marcelli, Paolo Napoletano, G. D. Fiore","doi":"10.1109/CAMP.2005.10","DOIUrl":"https://doi.org/10.1109/CAMP.2005.10","url":null,"abstract":"We present an architectural model for the interaction between top-down, object-based information and bottom-up, spatial-based information in determining visual attention shifts. We focus in particular on how the attentive process can take into account the processing of faces and multiple moving objects. To validate the model, experiments with eye-tracked human subjects are presented and discussed.","PeriodicalId":393875,"journal":{"name":"Seventh International Workshop on Computer Architecture for Machine Perception (CAMP'05)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134082538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The streaming computation model is appropriate for imaging applications because of the compute-intensive characteristics and memory access patterns. This paper advocates the streaming computation model and describes a streaming I/O peripheral used in a system-on-chip architecture. The impact on applications is described with details on performance gains from efficient memory transfers. Discussions on algorithm enhancements using streaming I/O peripherals are also provided.
{"title":"Streaming I/O for imaging applications","authors":"S. Chai, A. López-Lagunas","doi":"10.1109/CAMP.2005.50","DOIUrl":"https://doi.org/10.1109/CAMP.2005.50","url":null,"abstract":"The streaming computation model is appropriate for imaging applications because of the compute-intensive characteristics and memory access patterns. This paper advocates the streaming computation model and describes a streaming I/O peripheral used in a system-on-chip architecture. The impact on applications is described with details on performance gains from efficient memory transfers. Discussions on algorithm enhancements using streaming I/O peripherals are also provided.","PeriodicalId":393875,"journal":{"name":"Seventh International Workshop on Computer Architecture for Machine Perception (CAMP'05)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129436353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, a CMOS digital image sensor, which consists of A/D conversion, motion estimation circuits, and attention modules for ROI (region of interest) perception is presented. The functions of A/D conversion and motion estimation are implemented by 0.6 m CMOS processing circuit on chip, and attention modules are implemented off the chip as software currently. Attention modules are taken to improve limited applications of the smart image sensor. If the attention module is integrated on a smart sensor, we can use this sensor for tracking some regions of interest without outputting all pixels. The current smart image sensor responses to the changes of intensity, and uses the integration time to estimate motion. To make up for inherent property of the sensor from circuit design and extend its applications we decide to introduce some perceptive solutions to the image sensor. Attention modules for still and moving images are employed to achieve such purposes. The suggested approach also makes the smart image sensor initiate perceptive functions for such cases that motion estimation or intensity changes are not observed. Experimental results present the possibility that the smart image sensor can extract ROIs and use them when it selectively outputs pixels.
{"title":"A smart image sensor with attention modules","authors":"M. Park, K. Cheoi, T. Hamamoto","doi":"10.1109/CAMP.2005.7","DOIUrl":"https://doi.org/10.1109/CAMP.2005.7","url":null,"abstract":"In this paper, a CMOS digital image sensor, which consists of A/D conversion, motion estimation circuits, and attention modules for ROI (region of interest) perception is presented. The functions of A/D conversion and motion estimation are implemented by 0.6 m CMOS processing circuit on chip, and attention modules are implemented off the chip as software currently. Attention modules are taken to improve limited applications of the smart image sensor. If the attention module is integrated on a smart sensor, we can use this sensor for tracking some regions of interest without outputting all pixels. The current smart image sensor responses to the changes of intensity, and uses the integration time to estimate motion. To make up for inherent property of the sensor from circuit design and extend its applications we decide to introduce some perceptive solutions to the image sensor. Attention modules for still and moving images are employed to achieve such purposes. The suggested approach also makes the smart image sensor initiate perceptive functions for such cases that motion estimation or intensity changes are not observed. Experimental results present the possibility that the smart image sensor can extract ROIs and use them when it selectively outputs pixels.","PeriodicalId":393875,"journal":{"name":"Seventh International Workshop on Computer Architecture for Machine Perception (CAMP'05)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129786810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
S. Battiato, G. D. Blasi, G. Gallo, G. Messina, Salvatore Nicotra
The SVG (scalable vector graphics) standard allows representing complex graphical scenes by a collection of graphic vectorial-based primitives, offering several advantages with respect to classical raster images such as: scalability, resolution independence, etc. In this paper we present a full comparison between some advanced raster to SVG algorithms: SWaterG, SVGenie, SVGWave and some commercial tools. SWaterG works by a watershed decomposition coupled with some ad-hoc heuristics, SVGenie and SVGWave use a polygonalization based respectively on data dependent and wavelet triangulation. The results obtained by SWaterG, SVGenie and SVGWave are satisfactory both in terms of perceptual measured quality and compression ratio.
{"title":"SVG rendering for Internet imaging","authors":"S. Battiato, G. D. Blasi, G. Gallo, G. Messina, Salvatore Nicotra","doi":"10.1109/CAMP.2005.51","DOIUrl":"https://doi.org/10.1109/CAMP.2005.51","url":null,"abstract":"The SVG (scalable vector graphics) standard allows representing complex graphical scenes by a collection of graphic vectorial-based primitives, offering several advantages with respect to classical raster images such as: scalability, resolution independence, etc. In this paper we present a full comparison between some advanced raster to SVG algorithms: SWaterG, SVGenie, SVGWave and some commercial tools. SWaterG works by a watershed decomposition coupled with some ad-hoc heuristics, SVGenie and SVGWave use a polygonalization based respectively on data dependent and wavelet triangulation. The results obtained by SWaterG, SVGenie and SVGWave are satisfactory both in terms of perceptual measured quality and compression ratio.","PeriodicalId":393875,"journal":{"name":"Seventh International Workshop on Computer Architecture for Machine Perception (CAMP'05)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123059785","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
O. D. Robles, J. L. Bosque, L. Pastor, Angel Rodríguez
This paper presents a parallel implementation of a CBIR system which has been designed for several multiprocessor architectures. The paradigm that best fits the problem's needs is a farm based solution: a master process distributes the work load among the slave processes, and when they have finished, the master collects the partial results computed on each slave process. The multiprocessors used in the tests are a shared memory machine with eight processors - a Silicon Graphics Origin 2000 - and a cluster with 16 PC. In order to evaluate both architectures, the experimental results have been collected under several configurations, considering dynamic load balancing for the cluster based implementation. This paper analyzes the performance achieved by the different setups from the viewpoints of speed, price and scalability, presenting the conclusions that can be extracted from the results' comparison.
{"title":"Performance analysis of a CBIR system on shared-memory systems and heterogeneous clusters","authors":"O. D. Robles, J. L. Bosque, L. Pastor, Angel Rodríguez","doi":"10.1109/CAMP.2005.40","DOIUrl":"https://doi.org/10.1109/CAMP.2005.40","url":null,"abstract":"This paper presents a parallel implementation of a CBIR system which has been designed for several multiprocessor architectures. The paradigm that best fits the problem's needs is a farm based solution: a master process distributes the work load among the slave processes, and when they have finished, the master collects the partial results computed on each slave process. The multiprocessors used in the tests are a shared memory machine with eight processors - a Silicon Graphics Origin 2000 - and a cluster with 16 PC. In order to evaluate both architectures, the experimental results have been collected under several configurations, considering dynamic load balancing for the cluster based implementation. This paper analyzes the performance achieved by the different setups from the viewpoints of speed, price and scalability, presenting the conclusions that can be extracted from the results' comparison.","PeriodicalId":393875,"journal":{"name":"Seventh International Workshop on Computer Architecture for Machine Perception (CAMP'05)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117297618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
T. Komuro, S. Kagami, M. Ishikawa, Yoshio Katayama
An image sensor in which each pixel has a processing element is called a vision chip. The vision chip can perform real-time visual processing at a high frame rate of 1000 fps, and is expected to find application to intelligent robots and electronic devices. The vision chip has a different structure from existing processors and required special programming skills. This paper describes the development example of a vision chip compiler that enables easy programming of the vision chip.
{"title":"Development of a bit-level compiler for massively parallel vision chips","authors":"T. Komuro, S. Kagami, M. Ishikawa, Yoshio Katayama","doi":"10.1109/CAMP.2005.21","DOIUrl":"https://doi.org/10.1109/CAMP.2005.21","url":null,"abstract":"An image sensor in which each pixel has a processing element is called a vision chip. The vision chip can perform real-time visual processing at a high frame rate of 1000 fps, and is expected to find application to intelligent robots and electronic devices. The vision chip has a different structure from existing processors and required special programming skills. This paper describes the development example of a vision chip compiler that enables easy programming of the vision chip.","PeriodicalId":393875,"journal":{"name":"Seventh International Workshop on Computer Architecture for Machine Perception (CAMP'05)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121325037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
J. Denoulet, Ghilès Mostafaoui, L. Lacassagne, A. Mérigot
We present a robust implementation of a motion detection algorithm based on a Markovian relaxation both on general purpose processors, and on a specialized architecture, the associative mesh. The mesh architecture is an instance of the associative nets model targeting real time execution of low level image algorithms and vision-SoC implementation. The algorithm implementation on both architectures is described, and also the required optimizations to speedup the execution.
{"title":"Implementing motion Markov detection on general purpose processor and associative mesh","authors":"J. Denoulet, Ghilès Mostafaoui, L. Lacassagne, A. Mérigot","doi":"10.1109/CAMP.2005.31","DOIUrl":"https://doi.org/10.1109/CAMP.2005.31","url":null,"abstract":"We present a robust implementation of a motion detection algorithm based on a Markovian relaxation both on general purpose processors, and on a specialized architecture, the associative mesh. The mesh architecture is an instance of the associative nets model targeting real time execution of low level image algorithms and vision-SoC implementation. The algorithm implementation on both architectures is described, and also the required optimizations to speedup the execution.","PeriodicalId":393875,"journal":{"name":"Seventh International Workshop on Computer Architecture for Machine Perception (CAMP'05)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125191610","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A. Staiano, A. Ciaramella, L. D. Vinco, G. Longo, G. Raiconi, R. Tagliaferri, R. Amato, C. D. Mondo, G. Mangano, G. Miele
Due to the recent technological advances, data mining in massive data sets has evolved as a crucial research field for many if not all areas of research: from astronomy to high energy physics, to genetics etc. In this paper we discuss an implementation of the Probabilistic Principal Surfaces (PPS) which was developed within the framework of the AstroNeural collaboration. PPS are a nonlinear latent variable model which may be regarded as a complete mathematical framework to accomplish some fundamental data mining activities such as: visualization, clustering and classification of high dimensional data. The effectiveness of the proposed model is exemplified referring to a complex astronomical data set.
{"title":"Visualization, clustering and classification of multidimensional astronomical data","authors":"A. Staiano, A. Ciaramella, L. D. Vinco, G. Longo, G. Raiconi, R. Tagliaferri, R. Amato, C. D. Mondo, G. Mangano, G. Miele","doi":"10.1109/CAMP.2005.54","DOIUrl":"https://doi.org/10.1109/CAMP.2005.54","url":null,"abstract":"Due to the recent technological advances, data mining in massive data sets has evolved as a crucial research field for many if not all areas of research: from astronomy to high energy physics, to genetics etc. In this paper we discuss an implementation of the Probabilistic Principal Surfaces (PPS) which was developed within the framework of the AstroNeural collaboration. PPS are a nonlinear latent variable model which may be regarded as a complete mathematical framework to accomplish some fundamental data mining activities such as: visualization, clustering and classification of high dimensional data. The effectiveness of the proposed model is exemplified referring to a complex astronomical data set.","PeriodicalId":393875,"journal":{"name":"Seventh International Workshop on Computer Architecture for Machine Perception (CAMP'05)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125752572","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper describes the prey-predator multiagent system which can be considered as an abstraction of more complex real-world models. Both the prey and the predators are considered as autonomous agents with their own behaviors and perception of the environment. In particular, we propose a simulator which lets study different strategies such as cooperation and individualism. An extensive experiment has been carried out in order to prove the effectiveness of the latter.
{"title":"Prey-predator strategies in a multiagent system","authors":"B. Lenzitti, D. Tegolo, Cesare Valenti","doi":"10.1109/CAMP.2005.42","DOIUrl":"https://doi.org/10.1109/CAMP.2005.42","url":null,"abstract":"This paper describes the prey-predator multiagent system which can be considered as an abstraction of more complex real-world models. Both the prey and the predators are considered as autonomous agents with their own behaviors and perception of the environment. In particular, we propose a simulator which lets study different strategies such as cooperation and individualism. An extensive experiment has been carried out in order to prove the effectiveness of the latter.","PeriodicalId":393875,"journal":{"name":"Seventh International Workshop on Computer Architecture for Machine Perception (CAMP'05)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125317746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}