Mostrando recursos 21 - 40 de 76,126

  1. Goldenstein: Analysis of Facial Expressions in American Sign Language

    Christian Vogler; Siome Goldenstein
    In the age of speech and voice recognition technologies, sign language recognition is an essential part of ensuring equal access for deaf people. To date, sign language recognition research has mostly ignored facial expressions that arise as part of a natural sign language discourse, even though they carry important grammatical and prosodic information. One reason is that tracking the motion and dynamics of expressions in human faces from video is a hard task, especially with the high number of occlusions from the signers ’ hands. In this paper, we present a 3D deformable model tracking system to address this problem....

  2. Learning issues in a multi-modal robot-instruction scenario

    J. J. Steil; F. Röthling; R. Haschke; H. Ritter
    Abstract — One of the challenges for the realization of future intelligent robots is to design architectures which make user instruction of work tasks by interactive demonstration effective and convenient. A key prerequisite for enhancement of robot learning beyond the level of low-level skill acquisition is situated multi-modal communication. Currently, most existing robot platforms still have to advance to make the development of an integrated learning architecture feasible. We report on the status of the Bielefeld GRAVIS-robot architecture that combines statistical methods, neural networks, and finite state machines into an integrated system for instructing grasping tasks by human-machine interaction. It...

  3. Statistical transliteration for English-Arabic cross language information retrieval

    Nasreen Abduljaleel; Leah S. Larkey
    Out of vocabulary (OOV) words are problematic for cross language information retrieval. One way to deal with OOV words when the two languages have different alphabets, is to transliterate the unknown words, that is, to render them in the orthography of the second language. In the present study, we present a simple statistical technique to train an English to Arabic transliteration model from pairs of names. We call this a selected n-gram model because a two-stage training procedure first learns which n-gram segments should be added to the unigram inventory for the source language, and then a second stage learns...

  4. A Requirements Elicitation Approach Based in Templates and Patterns?

    A. Durán Toro; B. Bernárdez Jiménez; A. Ruiz Cortés; M. Toro Bonilla
    Abstract One of the main problems of requirements elicitation is expressing customer requirements in a form that can be understood not only by requirements engineers but also by noncomputer professional customers and users. The usual choice for expressing elicited requirements is natural language, since it is frequently the only common language to all participants. Problems of natural language are well–known, but using more formal notations too early is a risky choice that can make requirements impossible to understand for customers and users. Moreover, requirements engineers do not usually have good writing skills, and sometimes semantically correct requirements, expressed in natural...

  5. Interpretation as Abduction

    Jerry R. Hobbs; Mark Stickel; Paul Martin
    Abduction is inference to the best explanation. In the TACITUS project at SRI we have developed an approach to abductive inference, called “weighted abduction”, that has resulted in a significant simplification of how the problem of interpreting texts is conceptualized. The interpretation of a text is the minimal explanation of why the text would be true. More precisely, to interpret a text, one must prove the logical form of the text from what is already mutually known, allowing for coercions, merging redundancies where possible, and making assumptions where necessary. It is shown how such “local pragmatics ” problems as reference...

  6. How to deal with wicked anaphora

    Dan Cristea; Oana-diana Postolache
    This paper revises a framework (called AR-engine) capable of easily defining and operating models of anaphora resolution. The proposed engine envisages the linguistic and semantic entities involved in the cognitive process of anaphora resolution as represented in three layers: the referential expressions layer, the projected layer of referential expression’s features and the semantic layer of discourse entities. Within this framework, cases of anaphora resolution usually considered difficult to be tackled are investigated and solutions are proposed. Among them, one finds relations triggered by syntactic constraints, lemma and number disagreement, and bridging anaphora. The investigation uses a contiguous text from the...

  7. The use of locative expressions in dependence of the spatial relation between target and reference object in two-dimensional layouts

    Hubert D. Zimmer; Harry R. Speiser; Jörg Baus; Anselm Blocher
    Abstract. In two experiments we investigated the use of German locative expressions as a function of the spatial relation between a reference object (RO) and a to-be-located object (LO). In the experiments, a speaker described to another participant, by locative expressions, where LO can be found in relation to RO. LO (a blue dot) was presented at different positions around RO (a red dot). The listener saw RO only, and her or his task was to find LO by moving a small window over the screen using the computer mouse. The positions of LO were circularly arranged around RO and...

  8. Dynamic Cognition 2 Continuous Dynamics in Real-Time Cognition

    Michael Spivey; Michael J. Spivey; Rick Dale
    Rather than a sequence of logical operations performed on discrete symbols, real-time cognition is better described as continuously changing patterns of neuronal activity. The continuity in these dynamics indicates that, in between describable states of mind, much of our mental activity does not lend itself to the linguistic labels relied on by much of psychology. We discuss eye-tracking and mouse-tracking evidence for this temporal continuity, and provide geometric visualizations of mental activity depicted as a continuous trajectory through a neuronal state space. Close visitations of labeled attractors may constitute word recognition events and object recognition events, but the majority of...

  9. Mathematical Aspects of Command Relations


    In gb, the importance of phrase-structure rules has dwindled in favour of nearness conditions. Today, nearness conditions play a major role in defining the correct linguistic representations. They are expressed in terms of special binary relations on trees called command relations. Yet, while the formal theory of phrase-structure grammars is quite advanced, no formal investigation into the properties of command relations has been done. We will try to close this gap. In particular, we will study the intrinsic properties of command relations as relations on trees as well as the possibility to reduce nearness conditions expressed by command relations to...

  10. The Mathematics of Language

    Marcus Kracht
    Mathematical linguistics is concerned with the study of mathematical properties of natural languages and linguistic theories. Since the mathematical properties of interest to mathematical linguists are usually from theoretical computer science (complexity classes, language hierarchies, formal learnability), mathematical linguistics can be considered as an area of theoretical computational linguistics. However, since statistical methods are rarely used in mathematical linguistics, its relationship to current practices in computational linguistics is somewhat limited. While the introduction of logic in linguistic research has originally come from semantics, this line of work did not really use sophisticated meta-results. One of the main developments in mathematical...

  11. Linguistic Support for Large-Scale Distributed Programming

    Patrick Th; Eugster Rachid; Guerraoui Christian; H. Damm
    This paper presents linguistic primitives for publish/subscribe programming using events and objects. We integrate our primitives into a strongly typed objectoriented language through four mechanisms: (1) serialization,(2) multiple subtyping,(3) closures,and (4) deferred code evaluation. We illustrate our primitives through Java,showing how we have overcome its respective lacks. A precompiler transforms statements based on our publish/subscribe primitives into calls to specifically generated typed adapters,which resemble the typed stubs and skeletons generated by the rmic precompiler for remote method invocations in Java. 1

  12. Scalable Deep Linguistic Processing: Mind the Lexical Gap

    Timothy Baldwin
    Coverage has been a constant thorn in the side of deployed deep linguistic processing applications, largely because of the difficulty in constructing, maintaining and domain-tuning the complex lexicons that they rely on. This paper reviews various strands of research on deep lexical acquisition (DLA), i.e. the (semi-)automatic creation of linguistically-rich language resources, particularly from the viewpoint of DLA for precision grammars. 1

  13. Incorporating Syntactic Constraints in Recognizing Handwritten Sentences

    Rohini K. Srihari; Charlotte M. Baltus
    The output of handwritten word recognizers (HWR) tends to be very noisy due to various factors. In order to compensate for this behaviour, several choices of the HWR must be initially considered. In the case of handwritten sentence/phrase recognition, linguistic constraints may be applied in order to improve the results of the HWR. This paper discusses two statistical methods of applying syntactic constraints to the output of an HWR on input consisting of sentences/phrases. Both methods are based on syntactic categories (tags) associated with words. The rst is a purely statistical method, the second is a hybrid method which combines...

  14. Mutual Online Concept Learning for Multiple Agents


    To create multi-agent systems that are both adaptive and open, agents must collectively learn to generate and adapt their own concepts, ontologies, interpretations, and even languages actively in an online fashion. A central issue is the potential lack of any pre-existing concept to be learned; instead, agents may need to collectively design aconcept that is evolving as they exchange information. This paper presents a framework for mutual online concept learning (MOCL) in a shared world. MOCL extends classical online concept learning from single-agent to multi-agent settings. Based on the Perceptron algorithm, we present a specific MOCL algorithm, called the mutual...

  15. $rec.titulo


    This paper introduces a new theoretical formalism to specify private information. It utilizes single-referent linguistic assertions in defining ‘private information ’ in terms of ‘atomicity ’ and identification. Atomicity refers to the private information of a single person, in contrast to referring to compound information (e.g., X and Y are in love). Compound private information is reducible to atomic ones (X and someone are in love, and Y and someone are in love). The purpose of this reduction is to isolate “centers” of privacy for each person who is the proprietor of the information. We show that this discretion of...

  16. Language-Neutral Representation of Syntactic Structure


    We propose a semantically motivated linguistic representation called Language-Neutral Syntax (LNS), which is scalable in three important respects. First, though information stored in LNS is directly related to the surface tree, it is abstract enough to normalize many surface variations both within and across languages. Second, LNS is adaptable to new applications, in that any application that requires a particular kind of semantic information can extract that information from LNS. Finally, since LNS is developed as part of a broad-coverage parser, it is designed to handle a wide range of constructions. LNS is currently implemented in a large-scale, multi-lingual natural...

  17. Understanding Linguistic Evolution by Visualizing the Emergence

    Simon Kirby
    Abstract We show how cultural selection for learnability during the process of linguistic evolution can be visualized using a simple iterated learning model. Computational models of linguistic evolution typically focus on the nature of, and conditions for, stable states. We take a novel approach and focus on understanding the process of linguistic evolution itself. What kind of evolutionary system is this process? Using visualization techniques, we explore the nature of replicators in linguistic evolution, and argue that replicators correspond to local regions of regularity in the mapping between meaning and signals. Based on this argument, we draw parallels between phenomena...

  18. Face Translation: A Multimodal Translation Agent


    In this paper, we present Face Translation, a translation agent for people who speak different languages. The system can not only translate a spoken utterance into another language, but also produce an audio-visual output with the speaker’s face and synchronized lip movement. The visual output is synthesized from real images based on image morphing technology. Both mouth and eye movements are generated according to linguistic and social cues. An automatic feature extracting system can automatically initialize the system. After initialization, the system can generate synchronized visual output based on a few pre-stored images. The system is useful for a video...

  19. Clinical Forum Phonological Complexity and

    Language Learnability; Judith A. Gierut
    learnability to applications in clinical treatment of children with functional phonological delays. Method: The focus of the narrative review is on phonological complexity. This follows from learnability theory, whereby complexity in the linguistic input to children has been shown to trigger language learning. Drawing from the literature, phonological complexity is defined from epistemic, ontological, and functional perspectives, with specific emphasis on the application of Children’s acquisition of language occurs rapidly, with relatively few errors and seemingly without effort. In a matter of just about 36 months, a child typically produces novel sentences that involve complicated constructions,

  20. ABSTRACT Real-Time Computerized Annotation of Pictures ∗

    Jia Li; James Z. Wang
    Automated annotation of digital pictures has been a highly challenging problem for computer scientists since the invention of computers. The capability of annotating pictures by computers can lead to breakthroughs in a wide range of applications including Web image search, online picture-sharing communities, and scientific experiments. In our work, by advancing statistical modeling and optimization techniques, we can train computers about hundreds of semantic concepts using example pictures from each concept. The ALIPR (Automatic Linguistic Indexing of Pictures- Real Time) system of fully automatic and high speed annotation for online pictures has been constructed. Thousands of pictures from an Internet...

Aviso de cookies: Usamos cookies propias y de terceros para mejorar nuestros servicios, para análisis estadístico y para mostrarle publicidad. Si continua navegando consideramos que acepta su uso en los términos establecidos en la Política de cookies.