Brooks, R., Greiner, R., and Binford, T., "The ACRONYM Model-based Vision System," Proc. of IJCAI-6 , Tokyo, Japan, pp. 105--113, 1979.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....in [20] 1 Figure 1: Background Trees Modeled with Variable Radius Offset Surfaces. a) b) Figure 2: Generalized Cylinder as a Union of: a) Cross sectional Curves and (b) Profile Curves. 2 1. 1 Previous Work Generalized cylinder was first developed for the applications in computer vision [1, 9]. Using only a few simple procedural rules, the generalized cylinder can easily model a large class of 3D curved objects such as airplane, snake, horse, and doll [1, 9] Shani and Ballard [25] give a good review on early development of generalized cylinders in computer vision. The goal of modeling ....

....and (b) Profile Curves. 2 1. 1 Previous Work Generalized cylinder was first developed for the applications in computer vision [1, 9] Using only a few simple procedural rules, the generalized cylinder can easily model a large class of 3D curved objects such as airplane, snake, horse, and doll [1, 9]. Shani and Ballard [25] give a good review on early development of generalized cylinders in computer vision. The goal of modeling 3D curved objects in computer vision is, however, much different from that of computer graphics, CAD CAM, and computer animation. The 3D data modeling in computer ....

Brooks, R., Greiner, R., and Binford, T., "The ACRONYM Model-based Vision System," Proc. of IJCAI-6 , Tokyo, Japan, pp. 105--113, 1979.

....image understanding system which would solve many different classes of image analysis problems in a completely domain independent fashion. Not surprisingly, great efforts have been devoted to pursuing this goal. Computational theories have been developed[1, 2, 32] and computer systems built[4, 5, 31]. But in spite of this, the general vision systems have, overall, failed to compete with domain or problem specific systems in solving practical image analysis problems. Thus, on the one hand, it seems reasonable to expect that a useful image interpretation system has to be designed in a ....

Brooks,R. A., Greiner,R., and Binford,T. O., The ACRONYM Model-Based Vision System, Proc. of the International Joint Conference on Artificial Intelligence, 105113, 1979.

....that can be assembled in an infinite number of ways to cover situations that have never before been encountered. Model matching, along with large model bases representing objects, collections of objects, and geographical locales, undoubtedly plays a central role in fluent visual perception [Brooks et al. 79, Grimson 90, Hanson Riseman 78] But model matching alone is not sufficient to enable the understanding of novel scenes, or to enable the acquisition of new models. This requires the development of representation languages out of which meaningful models can be constructed. In sum, in order to ....

R. Brooks, R. Greiner, & T. Binford. The ACRONYM model-based vision system. In Proc. IJCAI, #6, Tokyo, Japan, pp. 105-113. 1979.

....vision. The problem is to come up with representations that are sophisticated enough to model interesting objects, yet simple enough to permit recognition or construction from images or other sensor data. A variety of surface based models have been proposed, including generalized cylinders [Brooks et al. 1979], superquadrics [Pentland, 1986] and deformable finite element models [Terzopoulos et al. 1987] Volumetric models such as octrees [Jackins and Tanimoto, 1980; Meagher, 1982] have also been used. One popular approach to constructing such volumes has been to intersect multiple silhouettes of ....

R. A. Brooks, R. Greiner, and T. O. Binford. The ACRONYM modelbased vision system. In Sixth International Joint Conference on Artificial Intelligence (IJCAI-79), pages 105--113, Tokyo, Japan, August 1979.

....next step in object recognition is the construction of the 2 1 2 D sketch, another representation in register with the image that contains depth and orientation information. Finally, a 3 D model is constructed of the object in the image, and that model is matched with stored 3 D models. Acronym [Brooks et al. 1979] is an object recognition system that follows Marr s approach quite closely. In the system, objects are modelled as generalized cylinders, and they are assumed to project as elongated ribbons in an image. Since the projections are elongated, the axis of the cylinder can be identified from the ....

....the other, or it may be stored as a number of samples or some other function. This idea was first introduced as the generalized cone by Binford [1971] current examples of the use of generalized cones and cylinders include work by Marr and Nishihara [1976] Agin [1977] Soroka and Bajcsy [1978] Brooks et al. 1979], and Shafer and Kanade [1983] One type of volumetric representation is the spatial occupancy representation. In this type of representation, the volume of space that an object occupies is represented by cutting up space up into a number of cubes or other cells and representing the particular ....

Brooks, R., R. Greiner, and T. Binford, "The ACRONYM model-based vision system," Proc. 6th Int. Conf. Artificial Intelligence, 1979, pp. 105-113.

No context found.

R.A. Brooks, G. Russell, and T. Binford. The acronym model based vision system. In Proceedings IJCAI, pages 105--113, 1979.

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