Waltz, D. (1975). Generating semantic descriptions from drawings of scenes with shadows. In P. Winston (Ed.), The psychology of computer vision (pp. 19--91). New York: McGraw-Hill.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....chain of Cartesian intersections, that has to be evaluated from the left to the right, will be written without parenthesis. For instance, AuBuCmeans uC. Remark 1 To . nd theC artesian domain [S] of A B C; by using only the operator u, a classical idea coming from constraint propagation [17] can be used: compute the fol lowing sequence ofC#WU2PP domains: S] 1) uA, uB, uC, S] 4) S] 3)uA, Because of the narrowing property (see (iv) formulae (7) the sequence [S] k) is always an enclosure of and converges (sometimes in a . nite numbers of steps) to aC ....

Waltz D. L., Generating semantic descriptions from drawings of scenes with shadows. in: P.H . Winston , ed i tor , The Psychology ofC omputer Vision, McGraw-Hill, New York, 1975, 19-91.

....presented in Section 4, which is based on interval constraint propagation (ICP) brie#y recalled in the next section. 3 Interval ari thmeti c andconstrai nt propagati on The approach to be employed combines two complementary tools, namely interval analysis [17] and constraint propagation [22] into what is known as interval constraint propagation (ICP) 6] 7] Note that interval analysis is also used for reliable global optimization without constraint propagation, see, e.g. 8] 23] in a general context and [24] 13] in a minimax context. Reliable global optimization based on ICP ....

D.L. Waltz. Generating semantic descriptions from drawings of scenes with shadows. In P.H. Winston, editor, The Psychology of Computer Vision, pages 19--91, New-York, 1975. McGraw-Hill.

....at the results, not only could I still see a chair without using color information, but I could also perceive the shape of the object without Correspondence to: T. Kanade, School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213, USA. E mail: kanade cs.cmu.edu. 0004 3702 93 06.00 1993 Elsevier Science Publishers B.V. All rights reserved T. Kanade Fig. 1. An office scene one of Ohlander s image set. Fig. 2. Ohlander s segmentation result. knowing it was a chair. In other words, I realized that there must exist geometrical constraints which enable us to recover ....

....to feel frustrated. One day, Allen Newell asked me to come to his office and talk about my research. I felt embarrassed about not having achieved much yet. When I finished explaining what I had been doing, Newell asked what the difference is between my method and Waltz s line labeling method [6] I gave a very vague answer, Waltz s method cannot handle this image, and mine will be more flexible . When I returned to my office, I quickly found the exact reason why Waltz s labeling method did not work with Fig. 3. Its trihedral world assumption (i.e. at each vertex three planes meet) ....

D. Waltz, Generating semantic descriptions from drawings of scenes with shadows, in: P.H. Winston, ed., The Psychology of Computer Vision (McGraw-Hill, New York, 1975).

....in the model (similar to [2] The next step is the automatic extraction of the constraints out of the scene. Few papers have dealt with the automatic extraction leaving it to the user to specify them [11, 14] The interpretation of the scene is formalised as a constraint satisfaction problem [13]. Liedtke used a semantic net for interpretation of architectural scenes [8] His interpretation is hypothesis driven. Hypotheses are verified or falsified by matching the 3D objects against the image. In our work we match the planes against a semantic net of a house by using a backtracking tree ....

D.L. Waltz. Generating Semantic Descriptions from Drawings of Scenes with Shadows. PhD thesis, AI Lab, MIT, 1972.

....in the model. The next step is the automatic extraction of the relationships between the extracted scene features. Few papers have dealt with the automatic extraction leaving it to the user to specify them [14, 17] The interpretation of the scene is formalised as a constraint satisfaction problem [16]. Liedtke used a semantic net for interpretation of architectural scenes [11] His interpretation is hypothesis driven. Hypotheses are verified or falsified by matching the 3D objects against the image. In our work we match the planes against a semantic net of a generic house by using a ....

D.L. Waltz. Generating Semantic Descriptions from Drawings of Scenes with Shadows. PhD thesis, AI Lab, MIT, 1972.

....are computationally dicult in general (i.e. they are NP hard) and require considerable expertise in optimization, software engineering, and the application domain. The constraint satisfaction approach to combinatorial optimization, which emerged from research in arti cial intelligence (e.g. [5, 15, 16, 18, 34]) and programming languages (e.g. 3, 28, 11] consists of a tree search algorithm using constraints to prune the search space at every node. Solving a problem in this approach amounts to specifying a set of constraints describing the solution set and a search procedure indicating how to search ....

D. Waltz. Generating Semantic Descriptions from Drawings of Scenes with Shadows. Technical Report AI271, MIT, MA, November 1972.

....enough to distinguish an object, it can be easily extended and refined later on by adding new constraints. For these reasons, constraints have been used for object modeling and for solving the matching problem of Computer Vision since the beginning of the study of Constraint Satisfaction [12]. The model of the object must be reliable and general: for this reason, it is often based on geometric and topological relations among features (see, for instance, 5, 7, 11] Each object can be represented by a constraint graph, where each characterizing primitive feature is a node and ....

D. I. Waltz, `Generating Semantic Descriptions from Drawings of Scenes with Shadows', in The Psychology of Computer Vision, ed., P. H. Winston, chapter 3, McGraw-Hill, New York, (1975).

....[10] ILOG Solver [28] and Numerica [33] address this problem by replacing floatingpoint arithmetic with interval arithmetic [3, 24] which guarantees the correctness of computations. The above mentioned solvers handle constraint systems by combining local consistency techniques and filtering [35, 23, 21]: a domain of possible values (interval) is associated to each variable; a constraint states a relation between some variables; solving a particular constraint then relies in discarding from some domains certain values for which the relation does not hold (inconsistency) This is done by applying ....

David L. Waltz. Generating semantic descriptions from drawings of scenes with shadows. In P. H. Winston, editor, The Psychology of Computer Vision. McGraw Hill, 1975.

....such as BNR Prolog, followed by clp(BNR) 6] Prolog IV [13] and Newton [5, 22] implement interval constraint solvers embedded in Prolog like languages. Essentially, the solvers combine local consistency techniques and filtering (propagation) two notions from Constraint Satisfaction Problems [24, 18, 17]. Intuitively, local application of operators narrowing down the domains of possible values for some variables, are iterated using local consistency techniques, until reaching a fixed point (no domain can be further reduced) The languages clp(BNR) and Prolog IV decompose complex constraints into ....

D. L. Waltz. Generating Semantic Descriptions from Drawings of Scenes with Shadows. In P. H. Winston, editor, The Psychology of Computer Vision. McGraw Hill, 1975.

....proposons d impl ementer un algorithme de propagation d intervalles bas e sur des m ethodes de recherche Tabou. Les r esultats exp erimentaux montrent une am elioration des temps de calcul pour un ensemble de tests propos es. 1 Introduction Les Probl emes de Satisfaction de Contraintes (CSPs) [17, 15, 16] mod elisent des taches exprim es par des relations (contraintes) entre des param etres (variables) auxquels sont associ es des ensembles de valeurs possibles (domaines) Le probl eme est de r eduire autant que possible les domaines en supprimant des valeurs ne satisfaisant pas les contraintes. La ....

....en supprimant des valeurs ne satisfaisant pas les contraintes. La r eduction des domaines est effectu ee par des m ethodes de consistance [15, 5] qui sont souvent des variations de la consistance d arc impl ement ees par des algorithmes de propagation de contraintes du type de celui de Waltz [17]. L id ee principale est de supprimer des valeurs inconsistantes des domaines a partir d un groupe (local) de contraintes et de propager les modifications aux contraintes contenant ces variables jusqu a ce qu il n y ait plus de r eduction possible. Davis [4] a introduit un type particulier de ....

[Article contains additional citation context not shown here]

D. L. Waltz. Generating Semantic Descriptions from Drawings of Scenes with Shadows. In P. H. Winston, editor, The Psychology of Computer Vision. McGraw Hill, 1975.

No context found.

Waltz, D. (1975). Generating semantic descriptions from drawings of scenes with shadows. In P. Winston (Ed.), The psychology of computer vision (pp. 19--91). New York: McGraw-Hill.

No context found.

Waltz, D. (1975). Generating semantic descriptions from drawings of scenes with shadows. In P. Winston (Ed.), The psychology of computer vision (pp. 19--91). New York: McGraw-Hill.

No context found.

D.L. Waltz. Generating semantic descriptions from drawings of scenes with shadows. Tech.Rep. MAC AI-271, MIT, 1972.

No context found.

D. L. Waltz. Generating semantic descriptions from drawings of scenes with shadows. Technical Report AI-TR-271, MIT Laboratory for Computer Science, Cambridge, MA, 1972.

No context found.

D.M. Waltz, "Generating semantic descriptions from drawings of scenes with shadows",The psychology of computer vision, pp 19-92, New York, 1972.

No context found.

Waltz,D., Generating Semantic Descriptions from Drawings of Scenes with Shadows, Tech. Rept. AI271, MIT, Cambridge, 1972

No context found.

D. L. Waltz. Generating semantic descriptions from drawings of scenes with shadows. Technical Report MAC-AI-TR-271, MIT, Cambridge, MA, 1972.

No context found.

D. L. Waltz, "Generating semantic descriptions from drawings of scenes with shadows ", MAC-AI-TR-217, MIT, 1972

No context found.

D.I. Waltz. Generating semantic descriptions from drawings of scenes with shadows. In P.H. Winston, editor, The Psychology of Computer Vision, chapter 3. McGraw-Hill, New York, 1975.

No context found.

D. Waltz. Generating semantic descriptions from drawing of scenes with shadows. In P.H. Winston, editor, Psychology of Computer Vision, pages 19-- 92. McGraw-Hill, New York, 1975.

No context found.

D.M.Waltz. Generating Semantic Descriptions from Drawings of Scenes with Shadows. Tech Rept AI-TR-271, M.I.T., Cambridge USA, 1972.

No context found.

D.L. Waltz. Generating Semantic Descriptions from Drawings of Scenes with Shadows. PhD thesis, AI Lab, MIT, 1972.

No context found.

D.L. Waltz. Generating Semantic Descriptions from Drawings of Scenes with Shadows. PhD thesis, AI Lab, MIT, 1972.

No context found.

D.M.Waltz, Generating Semantic Descriptions from Drawings of Scenes with Shadows, Tech Report AITR -271, M.I.T., Cambridge USA, 1972.

No context found.

D.L. Waltz. Generating Semantic Descriptions from Drawings of Scenes with Shadows. PhD thesis, AI Lab, MIT, 1972.

First 50 documents  Next 50

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