The Experts below are selected from a list of 360 Experts worldwide ranked by ideXlab platform

Michael S. Brown - One of the best experts on this subject based on the ideXlab platform.

  • raw Image reconstruction using a self contained srgb jpeg Image with small memory overhead
    International Journal of Computer Vision, 2018
    Co-Authors: Rang M. H. Nguyen, Michael S. Brown
    Abstract:

    Most camera Images are saved as 8-bit standard RGB (sRGB) compressed JPEGs. Even when JPEG compression is set to its highest quality, the encoded sRGB Image has been significantly processed in terms of color and tone manipulation. This makes sRGB–JPEG Images undesirable for many computer vision tasks that assume a direct relationship between pixel values and incoming light. For such applications, the RAW Image Format is preferred, as RAW represents a minimally processed, sensor-specific RGB Image that is linear with respect to scene radiance. The drawback with RAW Images, however, is that they require large amounts of storage and are not well-supported by many imaging applications. To address this issue, we present a method to encode the necessary data within an sRGB–JPEG Image to reconstruct a high-quality RAW Image. Our approach requires no calibration of the camera’s colorimetric properties and can reconstruct the original RAW to within 0.5% error with a small memory overhead for the additional data (e.g., 128 KB). More importantly, our output is a fully self-contained 100% compliant sRGB–JPEG file that can be used as-is, not affecting any existing Image workflow—the RAW Image data can be extracted when needed, or ignored otherwise. We detail our approach and show its effectiveness against competing strategies.

  • raw Image reconstruction using a self contained srgb jpeg Image with only 64 kb overhead
    Computer Vision and Pattern Recognition, 2016
    Co-Authors: Rang M. H. Nguyen, Michael S. Brown
    Abstract:

    Most camera Images are saved as 8-bit standard RGB (sRGB) compressed JPEGs. Even when JPEG compression is set to its highest quality, the encoded sRGB Image has been significantly processed in terms of color and tone manipulation. This makes sRGB-JPEG Images undesirable for many computer vision tasks that assume a direct relationship between pixel values and incoming light. For such applications, the RAW Image Format is preferred, as RAW represents a minimally processed, sensor-specific RGB Image with higher dynamic range that is linear with respect to scene radiance. The drawback with RAW Images, however, is that they require large amounts of storage and are not well-supported by many imaging applications. To address this issue, we present a method to encode the necessary metadata within an sRGB Image to reconstruct a high-quality RAW Image. Our approach requires no calibration of the camera and can reconstruct the original RAW to within 0:3% error with only a 64 KB overhead for the additional data. More importantly, our output is a fully selfcontained 100% complainant sRGB-JPEG file that can be used as-is, not affecting any existing Image workflow - the RAW Image can be extracted when needed, or ignored otherwise. We detail our approach and show its effectiveness against competing strategies.

Alina Madalina Popescu - One of the best experts on this subject based on the ideXlab platform.

  • automated instantiation of heterogeneous fast flow cpu gpu parallel pattern applications in clouds
    Parallel Distributed and Network-Based Processing, 2014
    Co-Authors: Suresh Boob, Horacio Gonzalezvelez, Alina Madalina Popescu
    Abstract:

    Parallel scientific workloads typically entail highly-customised software environments, involving complex data structures, specialised systems software, and even distinct hardware, where virtualisation is not necessarily supported by third-party providers. Considering the expansion of cloud computing in different domains and the development of different proprietary (e.g. Amazon Web Services, Azure) and open source cloud platforms (Eucalyptus, OpenStack, OpenNebula), users should arguably be able to automatically and seamlessly migrate their parallel workloads across cloud platforms using standardised virtual machines. However, even if it is easier to migrate the workload between nodes when the nodes have a similar configuration on the same platform, the transition between different platforms typically raises different issues such as vendor lock-in, portability, and interoperability. In this paper, we describe our work to automatically deploy a complex parallel software stack on heterogeneous hybrid cloud platforms. We have elastically deployed FastFlow - a C/C++ pattern-based programming framework for multi-/many-core and distributed platforms -- using virtual machines on both CPU and GPU-based architectures between heterogeneous virtualised platforms. Our approach relies on the standard Open Virtualization Format (OVF) in order to achieve a universal description of virtual appliances. Such a description is not only useful for elastically migrating and deploying, but also to determine the hardware/system software configuration needed switching to any new (cloud) Image Format. We have successfully evaluated our work using virtual machines based on VirtualBox and Amazon Web Services on local cluster and public cloud providers.

  • Automated Instantiation of Heterogeneous Fast Flow CPU/GPU Parallel Pattern Applications in Clouds
    2014 22nd Euromicro International Conference on Parallel Distributed and Network-Based Processing, 2014
    Co-Authors: Suresh Boob, Horacio González-vélez, Alina Madalina Popescu
    Abstract:

    Parallel scientific workloads typically entail highly-customised software environments, involving complex data structures, specialised systems software, and even distinct hardware, where virtualisation is not necessarily supported by third-party providers. Considering the expansion of cloud computing in different domains and the development of different proprietary (e.g. Amazon Web Services, Azure) and open source cloud platforms (Eucalyptus, OpenStack, OpenNebula), users should arguably be able to automatically and seamlessly migrate their parallel workloads across cloud platforms using standardised virtual machines. However, even if it is easier to migrate the workload between nodes when the nodes have a similar configuration on the same platform, the transition between different platforms typically raises different issues such as vendor lock-in, portability, and interoperability. In this paper, we describe our work to automatically deploy a complex parallel software stack on heterogeneous hybrid cloud platforms. We have elastically deployed FastFlow - a C/C++ pattern-based programming framework for multi-/many-core and distributed platforms -- using virtual machines on both CPU and GPU-based architectures between heterogeneous virtualised platforms. Our approach relies on the standard Open Virtualization Format (OVF) in order to achieve a universal description of virtual appliances. Such a description is not only useful for elastically migrating and deploying, but also to determine the hardware/system software configuration needed switching to any new (cloud) Image Format. We have successfully evaluated our work using virtual machines based on VirtualBox and Amazon Web Services on local cluster and public cloud providers.

Rang M. H. Nguyen - One of the best experts on this subject based on the ideXlab platform.

  • raw Image reconstruction using a self contained srgb jpeg Image with small memory overhead
    International Journal of Computer Vision, 2018
    Co-Authors: Rang M. H. Nguyen, Michael S. Brown
    Abstract:

    Most camera Images are saved as 8-bit standard RGB (sRGB) compressed JPEGs. Even when JPEG compression is set to its highest quality, the encoded sRGB Image has been significantly processed in terms of color and tone manipulation. This makes sRGB–JPEG Images undesirable for many computer vision tasks that assume a direct relationship between pixel values and incoming light. For such applications, the RAW Image Format is preferred, as RAW represents a minimally processed, sensor-specific RGB Image that is linear with respect to scene radiance. The drawback with RAW Images, however, is that they require large amounts of storage and are not well-supported by many imaging applications. To address this issue, we present a method to encode the necessary data within an sRGB–JPEG Image to reconstruct a high-quality RAW Image. Our approach requires no calibration of the camera’s colorimetric properties and can reconstruct the original RAW to within 0.5% error with a small memory overhead for the additional data (e.g., 128 KB). More importantly, our output is a fully self-contained 100% compliant sRGB–JPEG file that can be used as-is, not affecting any existing Image workflow—the RAW Image data can be extracted when needed, or ignored otherwise. We detail our approach and show its effectiveness against competing strategies.

  • raw Image reconstruction using a self contained srgb jpeg Image with only 64 kb overhead
    Computer Vision and Pattern Recognition, 2016
    Co-Authors: Rang M. H. Nguyen, Michael S. Brown
    Abstract:

    Most camera Images are saved as 8-bit standard RGB (sRGB) compressed JPEGs. Even when JPEG compression is set to its highest quality, the encoded sRGB Image has been significantly processed in terms of color and tone manipulation. This makes sRGB-JPEG Images undesirable for many computer vision tasks that assume a direct relationship between pixel values and incoming light. For such applications, the RAW Image Format is preferred, as RAW represents a minimally processed, sensor-specific RGB Image with higher dynamic range that is linear with respect to scene radiance. The drawback with RAW Images, however, is that they require large amounts of storage and are not well-supported by many imaging applications. To address this issue, we present a method to encode the necessary metadata within an sRGB Image to reconstruct a high-quality RAW Image. Our approach requires no calibration of the camera and can reconstruct the original RAW to within 0:3% error with only a 64 KB overhead for the additional data. More importantly, our output is a fully selfcontained 100% complainant sRGB-JPEG file that can be used as-is, not affecting any existing Image workflow - the RAW Image can be extracted when needed, or ignored otherwise. We detail our approach and show its effectiveness against competing strategies.

Forrest G Hall - One of the best experts on this subject based on the ideXlab platform.

  • boreas te 20 soils data over the nsa msa and tower sites in raster Format
    ORNL DAAC, 2000
    Co-Authors: Forrest G Hall, Hugo Veldhuis, David E Knapp
    Abstract:

    The BOREAS TE-20 team collected several data sets for use in developing and testing models of forest ecosystem dynamics. This data set was gridded from vector layers of soil maps that were received from Dr. Hugo Veldhuis, who did the original mapping in the field during 1994. The vector layers were gridded into raster files that cover the NSA-MSA and tower sites. The data are stored in binary, Image Format files. The data files are available on a CD-ROM (see document number 20010000884), or from the Oak Ridge National Laboratory (ORNL) Distributed Active Center (DAAC).

  • boreas regional dem in raster Format and aeac projection
    ORNL DAAC, 2000
    Co-Authors: David E Knapp, Kristine Verdin, Forrest G Hall
    Abstract:

    This data set is based on the GTOPO30 Digital Elevation Model (DEM) produced by the United States Geological Survey EROS Data Center (USGS EDC). The BOReal Ecosystem-Atmosphere Study (BOREAS) region (1,000 km x 1000 km) was extracted from the GTOPO30 data and reprojected by BOREAS staff into the Albers Equal-Area Conic (AEAC) projection. The pixel size of these data is 1 km. The data are stored in binary, Image Format files.

  • boreas regional soils data in raster Format and aeac projection
    ORNL DAAC, 2000
    Co-Authors: Bryan Monette, David E Knapp, Forrest G Hall, J E Nickeson
    Abstract:

    This data set was gridded by BOREAS InFormation System (BORIS) Staff from a vector data set received from the Canadian Soil InFormation System (CanSIS). The original data came in two parts that covered Saskatchewan and Manitoba. The data were gridded and merged into one data set of 84 files covering the BOREAS region. The data were gridded into the AEAC projection. Because the mapping of the two provinces was done separately in the original vector data, there may be discontinuities in some of the soil layers because of different interpretations of certain soil properties. The data are stored in binary, Image Format files.

  • boreas forest cover data layers of the nsa in raster Format
    ORNL DAAC, 2000
    Co-Authors: Forrest G Hall, David E Knapp, Manning Tuinhoff
    Abstract:

    This data set was processed by BORIS staff from the original vector data of species, crown closure, cutting class, and site classification/subtype into raster files. The original polygon data were received from Linnet Graphics, the distributor of data for MNR. In the case of the species layer, the percentages of species composition were removed. This reduced the amount of inFormation contained in the species layer of the gridded product, but it was necessary in order to make the gridded product easier to use. The original maps were produced from 1:15,840-scale aerial photography collected in 1988 over an area of the BOREAS NSA MSA. The data are stored in binary, Image Format files and they are available from Oak Ridge National Laboratory. The data files are available on a CD-ROM (see document number 20010000884).

  • boreas forest cover data layers over the ssa msa in raster Format
    ORNL DAAC, 2000
    Co-Authors: Jaime Nickeson, Fern Gruszka, Forrest G Hall
    Abstract:

    This data set, originally provided as vector polygons with attributes, has been processed by BORIS staff to provide raster files that can be used for modeling or for comparison purposes. The original data were received as ARC/INFO coverages or as export files from SERM. The data include inFormation on forest parameters for the BOREAS SSA-MSA. Most of the data used for this product were acquired by BORIS in 1993; the maps were produced from aerial photography taken as recently as 1988. The data are stored in binary, Image Format files.

Chunqiang Tang - One of the best experts on this subject based on the ideXlab platform.

  • FVD: a High-Performance Virtual Machine Image Format for Cloud. This is the longer version of the USENIX’11 paper with the same title, available at https://researcher.ibm.com/ researcher/view project.php?id=1852
    2013
    Co-Authors: Chunqiang Tang
    Abstract:

    Fast Virtual Disk (FVD) is a new virtual machine (VM) Image Format and the corresponding block device driver developed for QEMU. QEMU does I/O emulation for multiple hypervisors, including KVM, Xen-HVM, and VirtualBox. FVD is a holistic solution for both Cloud and non-Cloud environments. Its feature set includes flexible configurability, storage thin provisioning without a host file system, compact Image, internal snapshot, encryption, copy-on-write, copy-on-read, and adaptive prefetching. The last two features enable instant VM creation and instant VM migration, even if the VM Image is stored on direct-attached storage. As its name indicates, FVD is fast. Experiments show that the throughput of FVD is 249 % higher than that of QCOW2 when using the Post-Mark benchmark to create files.

  • fvd a high performance virtual machine Image Format for cloud
    USENIX Annual Technical Conference, 2011
    Co-Authors: Chunqiang Tang
    Abstract:

    Fast Virtual Disk (FVD) is a new virtual machine (VM) Image Format and the corresponding block device driver developed for QEMU. QEMU does I/O emulation for multiple hypervisors, including KVM, Xen-HVM, and VirtualBox. FVD is a holistic solution for both Cloud and non-Cloud environments. Its feature set includes flexible configurability, storage thin provisioning without a host file system, compact Image, internal snapshot, encryption, copy-on-write, copy-on-read, and adaptive prefetching. The last two features enable instant VM creation and instant VM migration, even if the VM Image is stored on direct-attached storage. As its name indicates, FVD is fast. Experiments show that the throughput of FVD is 249% higher than that of QCOW2 when using the Post-Mark benchmark to create files.