In this article we propose an architecture of a tracking system that can judge its own performance by an auto-critical function. Performance drops can be detected which trigger an automatic parameter regulation module. This regulation module is an expert system that searches a parameter setting with better performance and returns it to the tracking system. With such an architecture, a robust tracking system can be implemented which automatically adapts its parameters in case of changes in the environmental conditions. This article opens a way to self-adaptive systems in detection and recognition.
{"title":"Automatic parameter regulation for a tracking system with an auto-critical function","authors":"Daniela Hall","doi":"10.1109/CAMP.2005.14","DOIUrl":"https://doi.org/10.1109/CAMP.2005.14","url":null,"abstract":"In this article we propose an architecture of a tracking system that can judge its own performance by an auto-critical function. Performance drops can be detected which trigger an automatic parameter regulation module. This regulation module is an expert system that searches a parameter setting with better performance and returns it to the tracking system. With such an architecture, a robust tracking system can be implemented which automatically adapts its parameters in case of changes in the environmental conditions. This article opens a way to self-adaptive systems in detection and recognition.","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":"126796139","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}
D. Ginhac, E. Prasetyo, M. Paindavoine, B. Heyrman
This paper describes the main principles of a vision sensor dedicated to the detecting and tracking faces in video sequences. For this purpose, a current mode CMOS active sensor has been designed using an array of pixels that are amplified by using current mirrors of column amplifier. This circuit is simulated using Mentor Graphics/spl trade/ software with parameters of a 0.6 /spl mu/m CMOS process. The circuit design is added with a sequential control unit which purpose is to realise capture of subwindows at any location and any size in the whole image.
{"title":"Principles of a CMOS sensor dedicated to face tracking and recognition","authors":"D. Ginhac, E. Prasetyo, M. Paindavoine, B. Heyrman","doi":"10.1109/CAMP.2005.43","DOIUrl":"https://doi.org/10.1109/CAMP.2005.43","url":null,"abstract":"This paper describes the main principles of a vision sensor dedicated to the detecting and tracking faces in video sequences. For this purpose, a current mode CMOS active sensor has been designed using an array of pixels that are amplified by using current mirrors of column amplifier. This circuit is simulated using Mentor Graphics/spl trade/ software with parameters of a 0.6 /spl mu/m CMOS process. The circuit design is added with a sequential control unit which purpose is to realise capture of subwindows at any location and any size in the whole image.","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":"132444729","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}
We report on a computational key-project in astrophysics. The project is aimed at studying the interaction of a supernova shock wave with interstellar clouds. We describe the numerical code used, namely FLASH, a multi-dimensional astrophysical hydrodynamics code for parallel computers developed at the FLASH center (The University of Chicago); our team collaborates with, and contributes to, the FLASH project. We discuss the resources required for the whole project, the I/O management, the performance and the scalability of the code on IBM/Sp4 at CINECA. Finally, we present a selection of results.
{"title":"3-dimensional hydrodynamic interaction of a supernova remnant shock with an isolated cloud","authors":"S. Orlando, G. Peres, F. Reale","doi":"10.1109/CAMP.2005.55","DOIUrl":"https://doi.org/10.1109/CAMP.2005.55","url":null,"abstract":"We report on a computational key-project in astrophysics. The project is aimed at studying the interaction of a supernova shock wave with interstellar clouds. We describe the numerical code used, namely FLASH, a multi-dimensional astrophysical hydrodynamics code for parallel computers developed at the FLASH center (The University of Chicago); our team collaborates with, and contributes to, the FLASH project. We discuss the resources required for the whole project, the I/O management, the performance and the scalability of the code on IBM/Sp4 at CINECA. Finally, we present a selection of results.","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":"132847653","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, we present a system generating free-viewpoint video in real-time using multiple cameras and a PC cluster. Our system firstly reconstructs a shape model of objects by the visual cone intersection method, secondly transforms the shape model represented in terms of a voxel form into a triangular patch form, thirdly colors vertexes of triangular patches, and finally displays the shape-color model from the virtual viewpoint directed by a user. In this paper, we describe implementation details of our system and show some experimental results.
{"title":"Implementing a real-time free-viewpoint video system on a PC-cluster","authors":"M. Ueda, Daisaku Arita, R. Taniguchi","doi":"10.1109/CAMP.2005.30","DOIUrl":"https://doi.org/10.1109/CAMP.2005.30","url":null,"abstract":"In this paper, we present a system generating free-viewpoint video in real-time using multiple cameras and a PC cluster. Our system firstly reconstructs a shape model of objects by the visual cone intersection method, secondly transforms the shape model represented in terms of a voxel form into a triangular patch form, thirdly colors vertexes of triangular patches, and finally displays the shape-color model from the virtual viewpoint directed by a user. In this paper, we describe implementation details of our system and show some experimental results.","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":"121749481","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}
One of the challenges of research in machine perception today is filling the gap between researchers and engineers, general algorithms and development of working systems. Key performance issues like robustness, adaptability, flexibility and real-time must be pursued together with simplicity of design. These requirements have accentuated the interest in multi-modular systems and data fusion. The literature has proposed two main paradigms of multi-modular fusion: redundancy (or combination) and selection (or switching). We compare the relative advantages and complementary uses of the two paradigms. Then we focus on selection, which seems to attract less attention than redundancy -at least in machine vision, where we operate. After summarizing our work on a selection-based fusion scheme named context commutation, we describe a methodology in ten steps to design selection-based systems. With this paper, we hope to contribute to the understanding of selection as an additional tool for researchers and engineers working on machine perception.
{"title":"On the role and design of selection-based perception","authors":"P. Lombardi, B. Zavidovique","doi":"10.1109/CAMP.2005.36","DOIUrl":"https://doi.org/10.1109/CAMP.2005.36","url":null,"abstract":"One of the challenges of research in machine perception today is filling the gap between researchers and engineers, general algorithms and development of working systems. Key performance issues like robustness, adaptability, flexibility and real-time must be pursued together with simplicity of design. These requirements have accentuated the interest in multi-modular systems and data fusion. The literature has proposed two main paradigms of multi-modular fusion: redundancy (or combination) and selection (or switching). We compare the relative advantages and complementary uses of the two paradigms. Then we focus on selection, which seems to attract less attention than redundancy -at least in machine vision, where we operate. After summarizing our work on a selection-based fusion scheme named context commutation, we describe a methodology in ten steps to design selection-based systems. With this paper, we hope to contribute to the understanding of selection as an additional tool for researchers and engineers working on machine perception.","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":"121923036","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 contribution approaches the problem of scratch restoration in old movies as an optimisation problem. The functional based on statistical properties of the image around the scratch is optimised using an ad-hoc genetic algorithm. Given the large amount of computational time needed by genetic algorithms, a network of standard workstations with heterogeneous operating systems has been used. Each workstation in the network works on each scratch to perform the restoration, and a specific machine works as a root node with the task of distributing jobs on the network and adding the outputted restored scratches back into the image.
{"title":"Restoration of vertical line scratches with a distributed genetic algorithm","authors":"F. Isgrò, D. Tegolo","doi":"10.1109/CAMP.2005.46","DOIUrl":"https://doi.org/10.1109/CAMP.2005.46","url":null,"abstract":"This contribution approaches the problem of scratch restoration in old movies as an optimisation problem. The functional based on statistical properties of the image around the scratch is optimised using an ad-hoc genetic algorithm. Given the large amount of computational time needed by genetic algorithms, a network of standard workstations with heterogeneous operating systems has been used. Each workstation in the network works on each scratch to perform the restoration, and a specific machine works as a root node with the task of distributing jobs on the network and adding the outputted restored scratches back into the image.","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":"127426265","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}
G. Taffoni, D. Maino, C. Vuerli, G. Castelli, R. Smareglia, A. Zacchei, T. Ensslin, F. Pasian
Planck, the ESA satellite aimed at mapping the microwave sky through two complete sky surveys, will fly in 2007. It is a space mission that is extremely demanding in terms of computing power and data storage. PLANCK simulations mimic the whole mission starting from a virtual sky (ideal or "contaminated" by introducing several noise sources). Most of the related software is already installed at the two Planck DPCs (HFI and LFI). Its main goal is validation of the pipeline modules that will be used to build the data reduction pipelines during the operative phase of the mission. The software can also be profitably used to face the challenging computing power and data storage requirements and also during the pre-launch phase of the mission. Grid, the new technology that enables the distribution and sharing of computing power and data storage capacity over a geographical network, seems to be a promising answer to Planck data storage and processing needs. In the framework of the Italian, LCG based, Grid infrastructure at INFN, we managed to run a number of experiments aimed at defining an application specific environment for the PLANCK simulation software and at distributing on the Grid the produced simulated data. Such experiments have proved that the "gridification" of Planck pipelines is not only possible but is extremely convenient in terms of data processing speed and available data storage as is well illustrated in this paper.
{"title":"Enabling Grid technologies for simulating the Planck LFI simulated mission","authors":"G. Taffoni, D. Maino, C. Vuerli, G. Castelli, R. Smareglia, A. Zacchei, T. Ensslin, F. Pasian","doi":"10.1109/CAMP.2005.24","DOIUrl":"https://doi.org/10.1109/CAMP.2005.24","url":null,"abstract":"Planck, the ESA satellite aimed at mapping the microwave sky through two complete sky surveys, will fly in 2007. It is a space mission that is extremely demanding in terms of computing power and data storage. PLANCK simulations mimic the whole mission starting from a virtual sky (ideal or \"contaminated\" by introducing several noise sources). Most of the related software is already installed at the two Planck DPCs (HFI and LFI). Its main goal is validation of the pipeline modules that will be used to build the data reduction pipelines during the operative phase of the mission. The software can also be profitably used to face the challenging computing power and data storage requirements and also during the pre-launch phase of the mission. Grid, the new technology that enables the distribution and sharing of computing power and data storage capacity over a geographical network, seems to be a promising answer to Planck data storage and processing needs. In the framework of the Italian, LCG based, Grid infrastructure at INFN, we managed to run a number of experiments aimed at defining an application specific environment for the PLANCK simulation software and at distributing on the Grid the produced simulated data. Such experiments have proved that the \"gridification\" of Planck pipelines is not only possible but is extremely convenient in terms of data processing speed and available data storage as is well illustrated in this paper.","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":"116303596","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 fast development of innovative tools to create user friendly and effective multimedia libraries, services and environments requires novel concepts to support storage, annotation and retrieval of huge amounts of digital audiovisual data. This article presents a technique to tackle the first instance of the problem in visual digital archives: classification using generic semantic descriptions. As a case study, classification abilities inherent to some important MFEG-7 low-level visual descriptors are explored and quantified.
{"title":"Climbing the semantic ladder: towards semantic semi-automatic image annotation using MPEG-7 descriptor schemas","authors":"E. Izquierdo, A. Dorado","doi":"10.1109/CAMP.2005.16","DOIUrl":"https://doi.org/10.1109/CAMP.2005.16","url":null,"abstract":"The fast development of innovative tools to create user friendly and effective multimedia libraries, services and environments requires novel concepts to support storage, annotation and retrieval of huge amounts of digital audiovisual data. This article presents a technique to tackle the first instance of the problem in visual digital archives: classification using generic semantic descriptions. As a case study, classification abilities inherent to some important MFEG-7 low-level visual descriptors are explored and quantified.","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":"128666640","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 work a sub-symbolic technique for automatic, data driven language models construction is presented. Such a technique can be used to arrange a language-modelling module, which can be easily integrated in existing speech recognition architectures, such as the well-found HTK architecture. The proposed technique takes advantages from both the traditional LSA approach and from a novel application of a probability space metric known as "Hellinger's distance". Experimental trials are also presented, in order to validate the proposed approach.
{"title":"A sub-symbolic approach to word modelling for domain specific speech recognition","authors":"F. Agostaro, G. Pilato, G. Vassallo, S. Gaglio","doi":"10.1109/CAMP.2005.8","DOIUrl":"https://doi.org/10.1109/CAMP.2005.8","url":null,"abstract":"In this work a sub-symbolic technique for automatic, data driven language models construction is presented. Such a technique can be used to arrange a language-modelling module, which can be easily integrated in existing speech recognition architectures, such as the well-found HTK architecture. The proposed technique takes advantages from both the traditional LSA approach and from a novel application of a probability space metric known as \"Hellinger's distance\". Experimental trials are also presented, in order to validate the proposed approach.","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":"123974098","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}
Jacques-Olivier Klein, L. Lacassagne, H. Mathias, S. Moutault, A. Dupret
In this paper, a programmable analog retina is presented and compared with state of the art MPU for embedded imaging applications. The comparison is based on the energy requirement to implement the same image processing task. Results showed that analog processing requires lower power consumption than digital processing. In addition, the execution time is shorter since the size of the retina is reasonably large.
{"title":"Low power image processing: analog versus digital comparison","authors":"Jacques-Olivier Klein, L. Lacassagne, H. Mathias, S. Moutault, A. Dupret","doi":"10.1109/CAMP.2005.33","DOIUrl":"https://doi.org/10.1109/CAMP.2005.33","url":null,"abstract":"In this paper, a programmable analog retina is presented and compared with state of the art MPU for embedded imaging applications. The comparison is based on the energy requirement to implement the same image processing task. Results showed that analog processing requires lower power consumption than digital processing. In addition, the execution time is shorter since the size of the retina is reasonably large.","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":"123092253","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}