Workpackage descriptions. Workpackage number : 1 B3. Workpackage description Description of work
ii. iii. translation of captor’s parameter set into MPEG-4 Body Animation Parameters (BAPs); modelling and animation of ViSiCAST-VH (analogous to VRML H-anim) virtual humanoid in MPEG-4 context; encoding of BAPs. Development of appropriate sign language Description Schemes (DSs) for ViSiCAST-GML within the framework of the emerging MPEG-7 standard for a general purpose ‘Multimedia Content Description Interface’, investigating also different approaches allowing the derivation of the corresponding descriptors from monoscopic video sequences.
1-1 Direct-sign GML-sign
Workpackage descriptions. Workpackage 1 Analysis of the State of the Art and Development of Concepts Workpackage number : 1 Start date or starting event: Month 1 Participant number: AIT ONB SR IMAC SUL LMG NRM KVA TARX MECH MUS MOT Person-months per participant: 8 4 7 10 3 1 1 1 6 1 3 11 Participant number: SPAC ALI CC SIE GRAN IAT ZEUS SI CERT ICCS VALT TINC Person-months per participant: 3 1 4 8 1 5 15 8 6 7 14 5 Analysis of the State of the Art and Development of Concepts (Phase I) To provide high quality digital assets and services it is necessary to analyse the available content at partner sites. The important decision to take consists in defining what objects or surrogates might attract potential users or could be a basis for follow up added value processing. The nature of data will not only be of cultural & scientific type; goods available in museum shops have to be documented too to provide internet access to the museum shop. It might be useful not only store “primary” goods immediately for use of further processing, but also “eye catcher” suited to invoke a on demand process eventually doing a painting according to a theme offered in the REGNET – store (“painting on demand”). It should also be investigated if real goods, or replicates thereof will be offered. This means that the necessary logistics has to be set up properly (e.g. framing a painting). Having the world wide market in mind, even a mineral collection which was eventually a candidate for disposal might raise value via an auction over the internet. The objects to be used within REGNET differ in type, size, quality and quantity. In general two streams of actions have to be considered: retrospective digitisation and day to day cataloguing work. The first baseline for this task will be to differentiate the object type: Real objects (found in museums), Bibliographic objects (library, archives), Media objects (surrogates like images, films, etc), The documentation and digitisation plan should recognize the possibilities (financial, personnel support available) of a cultural institution as well as latest technology. During this task it should be investigated, if the establishment of a digitisation centre might be useful to serve a region using the REGNET service infrastructure.
Workpackage descriptions. Below we describe the contribution of Actions to the different WP's. The following coding is applied to denote tasks: TaskWx.Ay represents a task of Workpackage x performed by Action y. If the TaskWx.Ay has many subtasks they are denoted by their Number z : XxxxXx.Xx.Xx. The values of x and y can range from 0 to 4 where A0 denotes the action of the Coordinator. We write Task.Wx.All to denote a task performed by all partners.
Workpackage descriptions. This section contains a description of each workpackage. The person-months given refer to the researchers that are employed by the project; additional labour will be provided by the main investigators themselves. Workpackage number: WP1 — Virtual machine and cost model Start date: month 0 Person-months for each participant: 9 pm for participant 1, 4 pm for participant 2 Definition of virtual machine platform, formalization of cost model, and collection of ex- amples. This workpackage provides the basis for the research in the remaining packages. The overall goal of the project is to build a framework for Java-style downloadable bytecode equipped with checkable certificates regarding their usage of resources. The executable part of these will be bytecode for a specific virtual machine, to be defined in 1a. Claims about resource usage will refer to the cost model defined in 1b, which specifies the amounts of various kinds of resource that are consumed during execution. For the sake of connecting with current practice, we will employ a version of the Java Virtual Machine Language. We plan to investigate the suitability of Microsoft’s .NET intermediate language as an alternative platform at an early stage (1d). If JVML turns out to be too restrictive for our purposes then we could switch attention to .NET.
a. Define virtual machine platform. (2 months) Deliverables: internal technical document We will define a subset of JVML that is expressive enough to permit straightforward compilation of the high-level language to be defined in 3, yet small enough that to be tractable as an object of study, admitting proofs like those to be done in 4. Some modifications or simplifying assumptions may be required.
b. Formalise cost model. (6 months) Here we will define the amount of resource consumed by running a unit of mobile code. This will take the form of a semantics for our VM that delivers a cost along with the computed result, using previous work on formalisation of JVM (e.g. [51], [14], [44] and/or [19]) as a basis. Ultimately, many different kinds of resource are of interest; we hope that by considering two different concrete examples — heap space and num- ber of system calls — the elements of a general account will emerge. There may well be a division into time-like measures that are additive with respect to independent components versus space-like measures where the cost of performing two indepen- dent computations is the maximum of the costs of the components. I...
Workpackage descriptions. This section contains a description of each workpackage. The person-months given refer to the researchers that are employed by the project; additional labour will be provided by the main investigators themselves. Workpackage number: WP1 — Virtual machine and cost model Start date: month 1 Person-months for each participant: 12 pm for participant 1, 0 pm for participant 2 Definition of virtual machine platform, formalization of cost model, and collection of ex- amples. This workpackage provides the basis for the research in the remaining packages. The overall goal of the project is to build a framework for Java-style downloadable bytecode equipped with checkable certificates regarding their usage of resources. The executable part of these will be bytecode for a specific virtual machine, to be defined in 1a. Claims about resource usage will refer to the cost model defined in 1b, which specifies the amounts of various kinds of resource that are consumed during execution. For the sake of connecting with current practice, we will employ a version of the Java Virtual Machine Language. We plan to investigate the suitability of Microsoft’s .NET intermediate language as an alternative platform at an early stage (1d). If JVML turns out to be too restrictive for our purposes then we could switch attention to .NET.
a. Define virtual machine platform.
Workpackage descriptions. Workpackage title: Frameworks and Calculi for Dynamic Software Assembly Workpackage number: 1 Start date or starting event: month 0 Participant: GE-DISI HWU ICSTM TO-DI Person-months per participant: 22 5 11 5
Workpackage descriptions. Workpackage 1: Tasks for the Research Theory and Development Group on Smart Adaptive Systems (RTD SAS) Objectives Description of work
Workpackage descriptions. Workpackage 1 Description
Workpackage descriptions. Workpackage 1 Analysis of the State of the Art and Development of Concepts Workpackage number : 1 Start date or starting event: Month 1 Participant number: AIT ONB SR IMAC SUL LMG NRM KVA TARX MECH MUS MOT
Workpackage descriptions
