E. Andr, G. Herzog, T. Rist, "Natural Language Access to Visual Data: Dealing with Space and Movement", 1st Workshop on Logical Semantics of Time, Space and Movement in Natural Language, Toulouse, France, 1989.  Home/Search   Document Not in Database   Summary   Related Articles   Check

.... or organized with respect to the LO s: player 5: left RightPenaltyArea] in front player 7] player 7: between player 3 player 5] at player 3] near ball] near OutfieldSide1] player3: in RightHalfField] near MiddleLine] SBSZ87] RS88] Sch90a] Sch89] Hay90] Hay89] AHR89] and [HRA90] 36 cf. Section 3; the input for the complete visualization task is the temporally ordered sequence of those sets (plus elementary velocity restrictions) called the propositional elementary structure; the complete referent of a composed event proposition is constructed by ....

Elisabeth Andr'e , G. Herzog, and T. Rist. Natural language access to visual data: Dealing with space and movement. Bericht 63, SFB 314, VITRA, Universitat des Saarlandes, Saarbrucken, 1989.

....(t (move free ball) i (STOP (t (move free ball) TRIGGERED (t (have ball pl2 ball) i (SUCCEED (t (move free ball) Figure 2.5: An example of a course diagram representing a ball transfer event. Spatial relationship s between various objects are represented by relational tuples (Andr e, Herzog Rist 1989) of the form: rel name, subject, ref obj 1 , ref obj 2 , ref obj n , orientation ) where rel name is the spatial relationship between the object to be located, subject, according to the orientation) with relation to one or more reference objects, ref obj 1: n . 16 Adapted from Herzog, ....

Andr'e, E., Herzog, G. & Rist, T. (1989), Natural language access to visual data: Dealing with space and movement, in `1st Workshop on Logical Semantics of Time, Space and Movement in Natural Language', Toulouse, France.

.... computation of topological and projective relations (cf. Talmy, 1983; Herskovits, 1986 ] in 2D and 3D 1 The CSR 3D system (Computation of Spatial Relations in 3D Space) is part of the VITRA (VIsual TRAnslator) project which deals with the relationship between vision and natural language (cf. [Andr e et al. 1989]) The system starts from a geometric scene description, given by a vision system (cf. Koller, 1992] and proceeds to describe a recorded video sequence or to answer questions about the location of the objects involved in a cognitively plausible manner. LO a SLO x z RO z y x y SRO Figure ....

E. Andr'e, G. Herzog, and T. Rist. Natural language access to visual data: Dealing with space and movement. In F. Nef and M. Borillo, editors, Proc. of the 1 st Workshop of Logical Semantics of Time, Space and Movement in Natural Language. Herm`es, 1989.

....visual data in dynamic environments can be used in natural language descriptions ( Herzog et al. 89, Schirra et al. 87, Huang et al. 94] A model based approach is used for automatically generating 3D representations of the environment. This approach has been examined in a soccer domain ( Andr e et al. 89] and a dynamic traffic scene domain ( Schirra et al. 87] The model presented here depends on experiences gained in these domains. Our current work is related to problems which occur if an agent moves through real or synthetic environments. The agent s task during its movement is to ....

E. Andr'e, G. Herzog, and T. Rist. Natural Language Access to Visual Data: Dealing with Space and Movement. In: F. Nef and M. Borillo (eds.), Logical Semantics of Time, Space and Movement in Natural Language. Proc. of 1 st Workshop. Herm`es, 1989.

....VITRA has started in 1985 as part of the German special collaboration programme SFB 314, AI Knowledge Based Systems. It deals with the relations between speaking and seeing, and aims at a completely operational form of reference semantics for what is visually perceived. The system SOCCER (cf. Andr e et al. 89] constructed in VITRA demonstrates the computational link between visual perceptions and natural language. Here, we concentrate on the interaction between the speaker system SOCCER and its listener model ANTLIMA (cf. Schirra Stopp 93] mediated by the component ANTLIMA KOREF. 2 Three ....

E. Andr'e, G. Herzog, and T. Rist. Natural language access to visual data: dealing with space and movement. Bericht 63, SFB 314, VITRA, Universitat des Saarlandes, Saarbrucken, 1989.

....to be done incrementally in a kind of pipelining: a selection of already recognized events is verbalized simultaneously with further event recognition. Beside ACTIONS, the architecture of the system consists, broadly, of the Core System including three components (cf. Fig. 2, HSE 89] and [AHR89] the component for Event Recognition produces a set of propositions interpreting the given percepts as instances of spatial and spatio temporal relations the former essentially correspond to static spatial predicates like being to the left of , the latter to predicates of motion events, ....

Elisabeth Andr'e, G. Herzog, and T. Rist. Natural language access to visual data: Dealing with space and movement. Bericht 63, SFB 314, VITRA, Universitšat des Saarlandes, Saarbršucken, 1989.

.... grounded in visual perception, can be exploited for the definition of a reference semantics, that does not assign simple truth values to spatial predications, but instead introduces a measure of degrees of applicability that expresses the extent to which a spatial relation is applicable (cf. Andr e et al. 89] Since different degrees of applicability can be expressed by linguistic hedges, such as directly or more or less , more exact scene descriptions are possible. Furthermore, if an object configuration can be described by several spatial predications, the degree of applicability is used to ....

E. Andr'e, G. Herzog, and T. Rist. Natural Language Access to Visual Data: Dealing with Space and Movement. Report 63, Universitat des Saarlandes, SFB 314 (VITRA), Saarbrucken, 1989. Presented at the 1 st Workshop on Logical Semantics of Time, Space and Movement in Natural Language, Toulouse, France.

.... like uber, over , above , unter, under , below , vor, in front of , hinter, behind , links, left , rechts, right , neben, beside requires an orientation of space, i.e. a certain frame of reference, either intrinsic, extrinsic, or deictic, which assigns the vertical and horizontal axes [5], 35] In an intrinsic reference frame the orientation is given by an inherent organization of the reference object and the intrinsic axes of the RO are adopted. In the extrinsic use of a projective prepositions the orientation is determined by the position of a possibly imaginary observer, i.e. ....

....(add pp :head in :func location :identifier pp 1 :regent vp 1) in der linken Spielhaelfte . Figure 1: From visual data to a verbal description The reference semantic analysis of spatial prepositions leads to the term spatial relation as a target language independent meaning concept [5]. Conversely, the 4 definition and representation of the semantics of spatial relations is an essential condition for the synthesis of spatial reference expressions in natural language. Spatial relations can be defined by specifying conditions for object configurations, such as the distance ....

E. Andr'e, G. Herzog, and T. Rist. Natural Language Access to Visual Data: Dealing with Space and Movement. Report 63, Universitat des Saarlandes, SFB 314 (VITRA), Saarbrucken, 1989. Presented at the 1 st Workshop on Logical Semantics of Time, Space and Movement in Natural Language, Toulouse, France.

.... grounded in visual perception, can be exploited for the definition of a reference semantics, that does not assign simple truth values to spatial predications, but instead introduces a measure of degrees of applicability that expresses the extent to which a spatial relation is applicable (cf. Andre et al. 1989]) Since different degrees of applicability can be expressed by linguistic hedges, such as directly or more or less , more exact scene descriptions are possible. Furthermore, if an object configuration can be described by several spatial predications, the degree of applicability is used to ....

....and relation(s) for verbalization. In the context of the VITRA project, different classes of spatial relations have been examined in more detail. Wazinski [1993a] and Wazinski [1993b] are concerned with topological relations. Orientation dependent relations are treated in Andre et al. 1987a] and Andre et al. 1989]. Since the frame of reference is explicitly taken into account, the system is able to cope with the intrinsic, extrinsic, and deictic use of directional prepositions (cf. Retz Schmidt [1988] Recently, the algorithms developed so far have been generalized for 3 dimensional geometric ....

E. Andr e, G. Herzog, T. Rist. Natural Language Access to Visual Data: Dealing with Space and Movement. Report 63, Universitat des Saarlandes, SFB 314 (VITRA), Saarbrucken, 1989, Presented at the 1st Workshop on Logical Semantics of Time, Space and Movement in Natural Language, Toulouse, France.

....spatial reference expressions in natural language. Spatial relations can be defined by specifying conditions for object configurations, such as the distance between objects or the relative position of objects. In this sense, a spatial relation characterizes a class of object configurations (cf. Andr e et al. 89] 3.1 Computing the Elementary Spatial Relations The perceptual information available to the robot system is encoded in a 3 dimensional geometric model (cf. Fig. 2) which can be accessed by the natural language component. The basic meanings of spatial relations are defined with respect to ....

E. Andr e, G. Herzog, and T. Rist. Natural Language Access to Visual Data: Dealing with Space and Movement. Report 63, Universit at des Saarlandes, SFB 314 (VITRA), Saarbr ucken, 1989. Presented at the 1 st Workshop on Logical Semantics of Time, Space and Movement in Natural Language, Toulouse, France.

No context found.

E. Andr, G. Herzog, T. Rist, "Natural Language Access to Visual Data: Dealing with Space and Movement", 1st Workshop on Logical Semantics of Time, Space and Movement in Natural Language, Toulouse, France, 1989.

No context found.

E. Andr'e, G. Herzog, and T. Rist. Natural Language Access to Visual Data: Dealing with Space and Movement. In: F. Nef and M. Borillo (eds.), Proc. of the 1 st Workshop of Logical Semantics of Time, Space and Movement in Natural Language. Herm`es, 1989.

No context found.

E. Andr'e, G. Herzog, and T. Rist. Natural Language Access to Visual Data: Dealing with Space and Movement. In: F. Nef and M. Borillo (eds.), Logical Semantics of Time, Space and Movement in Natural Language. Proc. of 1 st Workshop. Herm`es, 1989.

No context found.

E. Andr'e, G. Herzog, and T. Rist. Natural Language Access to Visual Data: Dealing with Space and Movement. In: F. Nef and M. Borillo (eds.), Proc. of the 1 st Workshop of Logical Semantics of Time, Space and Movement in Natural Language. Herm`es, 1989.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC