D. Marr and H. K. Nishihara, "Representation and recognition of the spatial organization of three-dimensional shapes," Proc. R. Soc. of London, B, vol. 200, pp. 269--294, 1978.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....the above models is that they do not model the statistical variation among individuals and the effects of clothes on human shape. Thus, they may be used for human tracking or figure animation, but they are not appropriate for detecting people of various shapes and clothing. Marr and Nishihara [17] proposed a hierarchical 3D human model. At the highest level of the hierarchy, the body is modeled as a large extended cylinder, which is then resolved into small cylinders forming limbs and torso, and so on to fingers and toes. This hierarchical 20 representation is stable in the presence of ....

....enables us to decompose a shape such as a leg at the right position as shown in Fig. 2.1(c) Singh et al. s scheme restricts the cut to cross a symmetry axis in order to avoid short but undesirable cuts. However, robust computation of symmetry axes is difficult since from their very definitions [17, 114, 115] most axes are extremely sensitive to noise. In this thesis, the constraint on the salience of a part is used to replace the second requirement in the short cut rule in order to avoid the computation of symmetry axes. According to Hoffman and Singh s study [49] there are three factors that ....

[Article contains additional citation context not shown here]

D. Marr and H.K. Nishihara, "Representation and Recognition of the Spatial Organization of Three-Dimensional Shapes," Proc. Roy. Soc. B, B-200, pp. 269-294, 1977.

....plan model in 2, then discuss body part detectors learned by two competing algorithms, SVM and RVM, in 3. 4 presents our approach for learning and decoding body plans. Finally, 5 presents some results and discuss future work. 2 The Pictorial Structure of People In the work of Marr Nishihara [9] and many others [4,13] pictures of people are described geometrically as hierarchical assemblies of generalized cylinders and components. These early works represent the position of a part C relative to its parent P using adjunct relationships parameterized by C s position and ....

D. Marr and H.K. Nishihara. Representation and recognition of the spatial organization of three dimensional structure. Proceedings of the Royal Society of London B, 200:269--294.

....objects, and under different viewpoints. There is also evidence, such as from psychological experiments performed by Biederman [2] that a similar scheme is used by the human visual system. The use of simpler parts to describe more complex objects has a long history in computer vision [3] [11], 14] 16] 18] However, the use of part based representations in computer vision systems has been limited. This is, we believe, largely due to the difficulty of automatically computing part based descriptions from a real image: The part decomposition hierarchy is not given a priori, we must ....

D. Marr and H.K. Nishihara, "Representation and Recognition of the Spatial Organization of Three-Dimensional Shapes," Proc. Royal Physical Soc., pp. 269-294, 1977.

....about each part and about the object as a whole to support form analysis and computation. A driving consideration faced by all skeleton methods is the condition put forth by Marr and Nishihara that small changes to a boundarys form should cause only small changes to the forms representation [Marr78]. This rule insists that skeletons remain stable under boundary change, and considerable effort has been spent on modifying Blums MAT to exhibit this kind of stability [August99] Szkely96] Complicating this rule is the problem that some boundary perturbations actually are important; the growth of ....

Marr, D., and K. H. Nishihara, Representation and Recognition of the Spatial Organization of Three Dimensional Structure, Proceedings of the Royal Society of London, 200:269-294. (1978).

....plan model in 2, then discuss body part detectors learned by two competing algorithms, SVM and RVM, in 3. 4 presents our approach for learning and decoding body plans. Finally, 5 presents some results and discusses future work. 2 The Pictorial Structure of People In the work of Marr Nishihara [15] and others [10,19] people are described as hierarchical 3D assemblies of generalized cylinders and components. The position of a part C relative to its parent P is parametrized by C s position and angular orientation in P s cylindrical coordinate system. Each joint ....

D. Marr and H.K. Nishihara. Representation and recognition of the spatial organization of three dimensional structure. Proceedings of the Royal Society of London B, 200:269--294, 1978.

....criteria for recognition algorithms are presented first. Criteria for a Robust Recognition Algorithm Several central requirements for a successful recognition algorithm can be identified. These issues have previously been raised by researchers in computer vision and object recognition [5] [6] [7] however they are of importance for recognition tasks in general. A recognition algorithm should degrade gracefully with increasing noise in the sensory input. A recognition algorithm should degrade gracefully with increasing complexity in the recognition task. recognition algorithm should ....

Mart, D. and Nishihara, H.K. "Representation and Recognition of the Spatial Organization of Three-Dimensional Shapes", Proc. R. Soc. London B 200, pp. 269-294, 1978.

....elements or lines. In that case, component detection is relatively easy, but their relationships become complicated. Some older object recognition systems (e.g. 22] and at least one psychological theory [14] describe objects in terms of elongated volumes called generalized cylinders ( 15] [94]) The use of generalized cylinder primitives may be regarded as a compromise between the 13 conflicting requirements of generality and detectability of object components. The description of an object under this representation scheme is said to be object centered, in the sense that it involves ....

D. Mart and H. K. Nishihara. Representation and recognition of the spatial organization of three dimensional structure. Proceedings of the Rolal Societal of London B, 200:269- 294, 1978.

....such a representation is by decomposing the object into parts. Parts are extracted by cutting the object in concavities [17, 22, 43] and labeled according to their general shape. The labels, together with the spatial relationships between the parts, are used to identify the class of the ob ject [4, 6, 7, 26]. A second approach extracts the parts of the object that fulfill certain functions. The list of func tions is used to determine the object s class [16, 39, 47] Schemes that break objects into parts are insufficient to explain all the aspects of recognition for the following reasons. First, in ....

Marr D. and Nishihara, H.K., 1978. Representation and recognition of the spatial organization of threedimensional shapes. Proc. of the Royal 'ociety, London, B200, 269-294.

....criteria for recognition algorithms are presented first. Criteria for a Robust Recognition Algorithm Several central requirements for a successful recognition algorithm can be identified. These issues have previously been raised by researchers in computer vision and object recognition [51 [6] 71, however they are of importance for recognition tasks in general. A recognition algorithm should degrade gracefully with increasing noise in the sensory input. A recognition algorithm should degrade gracefully with increasing complexity in the recognition task. recognition algorithm ....

....in the second row of Table 2, and can be compared with the second row of Table 1. In this condition, the lexical stress information provides some additional constraint. In particular, the stress pattern substantially increases the percentage of the lexicon which is uniquely specifiable from 6 to 25 percent. Since stress information plays a larger role when word frequency is taken into account, this indicates that stress is important in differentiating certain common words [rom one another. Perhaps subsequent psychophysical investigation can test whether stress is important in human ....

[Article contains additional citation context not shown here]

Mart, D. and Nishihara, H.K. "Representation and Recognition of the Spatial Organization of Three-Dimensional Shapes", Proc. R. Soc. London B 200, pp. 269-294, 1978.

....2. Frequency areas of different vowels in the two dimensional space generated by the first two formants. Values in cps. From [3] For more complex concepts like bird it is perhaps more difficult to describe the underlying conceptual space. However, if something like Marr and Nishihara s [13] analysis of shapes is adopted, we can begin to see how such a space would appear. 8 Their scheme for des cribing biological forms uses hierarchies of cylinder like modelling primitives. Each cylinder is described by two coordinates (length and width) Cylinders are combined by determining the ....

Marr, David and H. K. Nishihara (1978): "Representation and recognition of the spatial organization of three-dimensional shapes", Proceedings of the Royal Society in London, B 200, pp. 269-294.

....discussion 7. 1 Overview of the conclusions Experimental results presented in the preceding three sections speak against the notion that object representations in the human visual system are three dimensional, object centered and viewpoint independent (as stipulated, e.g. by Marr and Nishihara [16]) Our subjects always reported perceiving the stimuli in full 3D during training (due to the kinetic depth effect and other sources of 3D information) and, in specially designed experiments, during testing. Nevertheless, the subjects consistently behaved as if they had failed to commit truly ....

D. Marr and H. K. Nishihara. Representation and recognition of the spatial organization of three dimensional structure. Proceedings of the Royal Society of ondon B, 200:269 294, 1978.

....based on some of the literature on human perception. Particular attention was paid to structural object recognition theory. According to this theory as objects are perceived they are decomposed into 3D set of primitives called geons, together with the skeleton structure connecting them (Figure 1) [1,6]. 2 Figure 1. Sample geons and objects made from them. We conducted three studies in which the effectiveness of using 3D shaped primitives (geons) as components in node link diagrams were evaluated. The results of the first two studies show that the use of geon primitives in node link diagrams ....

Marr, D., and Nishihara, H.K. (1978) Representation and recognition of the spatial organization of three-dimensional shapes. Proceedings of the Royal Society of London, B, 200, 269-294.

....In the present paper we report on a more rigorous experiment to test whether it is really the 3D aspect of our geon based diagrams that makes them more effective. 2. Geon theory Marr and Nishihara were the first to develop a reasonably complete structure based approach to object recognition [4]. In their theory an object can be decomposed into axes of generalized cones depending on the nature of detail required. Thus for example, to differentiate between a horse and a giraffe, the decomposition would result in an arrangement of axes of differing lengths representing generalized ....

Marr, D., and Nishihara, H.K. (1978) Representation and recognition of spatial organization of three-dimensional shapes, Proceedings of the Royal Society of London, B, 200, 269-294.

....identified by a recursive search of each of the connected networks of nodes. 3.3.2. Perceiving Canonical Shape Marr and Nishihara describe vision as a process that produces from images of the external world a description that is useful to the viewer and not cluttered by irrelevant information. [11]. Rose s nodal network is a necessary interim stage for identifying the presence of volumes but it is very cluttered by irrelevant information, making it difficult to access the general salient features of volumes directly from their nodal networks. Marr and Nishihara show that an object s surface ....

....a macroscopic scale between volumes. This progression from large amounts of low level data to a small amount of high level information is a property common to many models of natural perception processes [12] typified by Marr and Nishihara s work on the recognition of three dimensional shapes [11]. 3.4. Representation Rose s method of creating drawings is based on the constructive model of representation in young children described in section 2.2. To recapitulate the constructive model, Rose will create an equivalent world within the two dimensional medium, rather than a projection of ....

Marr, D & Nishihara, H. Representation and Recognition of the Spatial Organisation of Three-dimensional Shapes. Proceedings of the Royal Society of London. Vol. B. 200. pp. 269-294. 1978.

....results are presented within each section. 40 41 4. 1 Related Work The image subregion querying problem is closely related to object recognition, which has been studied for a long time by the computer vision community [Robe65] Some of the early work in object recognition and detection [MN78] suggests that geometric cues such as edge, surface, and depth information be identified before object recognition is attempted. Most of such object recognition systems compare the geometric features of the model with those of an image using various forms of search (some of which are ....

D. Marr and H. K. Nishihara. Representation and recognition of the spatial organization of three-dimensional shapes. Proc. R. Soc. Lond. B., 200:269--294, 1978.

....legs and functional evidence are successfully (a) Image (b) Matches (c) S map Figure 7: Desk B (a) Image (b) Matches (c) S map Figure 8: Desk C detected. Figure 7 c represents the recognized desk in an s map. The front part of a desk is rendered with the thin line and the small circles. Figure 8 displays another desk scene. The desk has a monitor on it. Figure 8 b illustrates detected significant surfaces and functional evidence that are represented as thicker lines. All four boundaries of the desk top are successfully recovered although rear boundaries of the desk are occluded by a ....

D. Marr and K. Nishihara. Representation and recognition of the spatial organization of threedimensional shapes. Proceedings of the Royal Society of London, B(200):269--294, 1977.

....can be represented in many different ways depending on the level of detail needed by a system. The idea of representing humans at a digital form goes way back in history. But in the late 1970s two main papers were published, which set the standard for future work. These were by Marr and Nishihara [106] in 1979 and one year later by Badler and Smoliar [11] The main idea was to represent a human by a number of geometric shapes, usually cylinders. During the last decade researchers have identified a problem with this type of models, namely that it is too static. Therefore deformable models have ....

D. Marr and H.K. Nishihara. Representation and Recognition of the Spatial Organization of Three-Dimensional Shapes. In Proc. R. Soc. B., volume 200, pages 269--294, 1978.

....to stored object descriptions. One of the most prominent models of this type is the Recognition by Components (RBC) theory of Biederman [3, 19] whose emphasis on representing an object by decomposing it into basic geometrical shapes is reminiscent of the scheme proposed by Marr and Nishihara [25]. RBC predicts that recognition of objects should be viewpoint invariant as long as the same structural description can be extracted from the different object views. In contrast to structural description models, the basic tenet of view or image based models is that objects are represented as ....

Marr, D. and Nishihara, H. K. (1978). Representation and recognition of the spatial organization of three-dimensional shapes. Proc. R. Soc. Lond. B Biol. Sci. 200, 269--294.

....vary as the viewpoint changes. An important problem in 3D object recognition is therefore how to accomplish view invariant recognition. The traditional approach is to recover the 3D shape and to construct structural descriptions of 3D volumetric primitives in the object centered coordinate frames ([14], for example) Nevertheless, due to the complexity of 3D shape processing, the reliable construction of volumetric primitives is often di cult and time consuming. Alternatively, recent computational studies have explored view based approaches, where 3D objects are recognized more directly from ....

D. Marr, H.K. Nishihara, Representation and recognition of the spatial organization of threedimensional shapes, Proc. Roy. Soc. London B 200 (1978) 269}294.

....way, as for example a coffeepot might consist of a body in the form of a truncated cone, with a handle in the form of a half torus and a spout in the form of a cylinder obliquely cut off at one end. This type of volume based description is exemplified by the work of Biederman [ 1987 ] and of Marr and Nishihara [ 1978 ] 1 . The alternative approach is to characterise the form of the outline of the object, for example by noting the variation in curvature across different positions. Examples of outline based approaches to two dimensional shape are those of Leyton [ 1988 ] and of Hoffman and Richards [ 1982 ] ....

D. Marr and H. K. Nishihara. Representation and recognition of the spatial organization of three dimensional shapes. Proceedings of the Royal Society of London, Series B, 200:269--294, 1978.

....that is sufficient for the task at hand, and which also accounts for certain otherwise puzzling phenomena. This sort of minimalist top down strategy has occasionally been used to advantage in designing computational models. Good examples are Newell s (1973) production system architecture, and Marr and Nishihara s (1976) SPASAR mechanism for rotating 3 D models into a canonical orientation in the process of recognition. Inasmuch as it is also an attempt to work out a set of basic operations which can be used to create a procedure for carrying out the task, it is very similar in spirit to Ullman s (1984) ....

Marr, D., and Nishihara, H.K. (1976). Representation and Recognition of Spatial Organization of Three-Dimensional Shapes, MIT A.I. Memo 377:1-57.

....through its region [10] Second, an orthogonal classification is that 1 The avid bird watcher will assign these the labels hawk and cormorant . 1 of primitive based approaches, where shape is viewed statically as a combination of a small number of components, e.g. generalized cylinders [42], or geons [3] versus process based approaches [38] where shape is explained dynamically or developmentally via a set of processes acting on a simpler shape. Returning to Blum s proposal of a sphere as a primitive basis for modelling object growth, this view has spawned a vast literature on the ....

D. Marr and K. H. Nishihara. Representation and recognition of the spatial organization of three dimensional structure. Proceedings of the Royal Society of London, B 200:269--294, 1978.

No context found.

D. Marr and H. K. Nishihara, "Representation and recognition of the spatial organization of three-dimensional shapes," Proc. R. Soc. of London, B, vol. 200, pp. 269--294, 1978.

No context found.

Marr, D., and Nishihara, H. K. 1978. Representation and recognition of the spatial organization of three-dimensional shapes. Proceedings of the Royal Society B 200:269--294.

No context found.

D. Marr and K. Nishihara. Representation and recognition of the spatial organization of three dimensional structure. Proceedings of the Royal Society of London, 200:269--294, 1978.

No context found.

D. Marr and K. H. Nishihara. Representation and Recognition of the Spatial Organization of Three Dimensional Structure. Proceedings of the Royal Society of London, B 200:269--294, 1978.

No context found.

Marr. D. and Nishihara, H.K., Representation and recognition of the spatial organization of three dimensional shapes, Proc. of Royal Soc. of London, 1978.

No context found.

D. Marr and K. H. Nishihara. Representation and Recognition of the Spatial Organization of Three Dimensional Structure. Proceedings of the Royal Society of London, B 200:269--294, 1978.

No context found.

D. Marr and H. Nishihara. Representation and recognition of the spatial organization of three-dimensional shapes. Royal Society of London, B 200:269--294, 1978.

No context found.

D. Marr, K. H. Nishihara, Representation and recognition of the spatial organization of three dimensional structure, Proceedings of the Royal Society of London 200 (1978) 269--294.

No context found.

D. Marr, K. Nishihara, Representation and recognition of the spatial organization of three-dimensional shapes. Proc. R. Soc. London B 200 (1978) 269--294.

No context found.

D. Marr and H.K. Nishihara. Representation and Recognition of the Spatial Organization of Three-Dimensional Shapes. In Proc. R. Soc. B., volume 200, pages 269--294, 1978.

No context found.

D. Marr and H.K. Nishihara. Representation and recognition of the spatial organization of three-dimensional shapes. Proceedings of the Royal Society of London, Series B, Biological Sciences, 200:269--294, 1978.

No context found.

D. Marr and H.K. Nishihara. Representation and recognition of the spatial organization of threedimensional shapes. Proceedings of the Royal Society of London, Series B, (200):269--294, 1978.

No context found.

Marr, D. and H.K. Nishihara, Representation and Recognition of the Spatial Organization of Three Dimensional Shapes, in Proceedings of the Royal Society . 1978. p. 200, 269-294.

No context found.

D. Marr and H. Nishihara. Representation and recognition of the spatial organization of three dimensional structure. In Proceedings of the Royal Society of London, 1978.

No context found.

Marr D, Nishihara HK. Representation and recognition of the spatial organization of three-dimensional shapes. Proceedings of the Royal Society of London, Series B 1978; 200: 269-94.

No context found.

Marr, D. and H.K. Nishihara, Representation and Recognition of the Spatial Organization of Three Dimensional Shapes, in Proceedings of the Royal Society . 1978. p. 200, 269-294.

No context found.

D. Marr and H. K. Nishihara. Representation and recognition of the spatial organization of three-dimensional shapes. In Proceedings of the Royal Society of London B, number 200, pages 269--294, 1978.

No context found.

D. Marr and H.K. Nishihara. Representation and recognition of the spatial organization of three-dimensional shapes. Proceedings of the Royal Society of London, Series B, Biological Sciences, 200:269--294, 1978. 81

No context found.

D. Marr and K. Nishihara. Representation and recognition of the spatial organization of threedimensional shapes. Proc. Royal Society, London, B-200:269--294, 1978.

No context found.

D. Marr and H. Nishihara. Representation and recognition of the spatial organization of three-dimensional shapes. In Proc. of the Royal Society of London, series B, volume 200, pages 269--294, February 1978.

No context found.

D. Marr, and H.K. Nishihara, K., Representation and recognition of the spatial organization of three-dimensional shapes, Proceedings of the Royal Society of London 200B, 1978, 269-294.

No context found.

D. Marr and H.K. Nishihara. Representation and recognition of the spatial organization of threedimensional shapes. Proceedings of the Royal Society of London, Series B, (200):269--294, 1978.

No context found.

D. Marr and H.K. Nishihara, "Representation and Recognition of the Spatial Organization of Three-Dimensional Shapes," Proc. Roy. Soc. B, B-200, pp. 269-294, 1977.

No context found.

Marr, D. and Nishihara, H.K. (1978) Representation and recognition of the spatial organization of three-dimensional shapes. Proc. R. Soc. Lond. Ser. B 200, 269 -- 294

No context found.

D. Marr and H.K. Nishihara. Representation and recognition of the spatial organization of three dimensional structure. Proceedings of the Royal Society of London B, 200:269--294, 1978.

No context found.

D.Marr and H.Nishihara, "Representation and recognition of the spatial organization of three-dimensional shapes", Proc. Roy. Soc. London, B200, pp269-294,1978

No context found.

Marr, D and Nishihara, K `Representation and recognition of the spatial organization of three-

No context found.

D. Marr and H.K. Nishihara, "Representation and recognition of the spatial organization of three-dimensional shapes". Proceedings of the Royal Society of London, 1978, B, 200, pp. 269-294.

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