R. Szeliski, R. Weiss, Robust shape recovery from occluding contours using a linear smoother, Int. J. Comput. Vis. 28 (1) (1998) 27--44.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....geometry to produce renderings. Although just three views were used, the recovered shape is close to the actual shape because the views were chosen to align with the boxy geometry of the object. A project in which a continuous stream of views was used to reconstruct object geometry is presented in [45, 44]; see also Fig. 2.4. A similar silhouette based technique was used to provide an approximate estimate of object geometry to improve renderings in the Lumigraph image based modeling and rendering system [16] In modeling from silhouettes, qualitatively better results can be obtained for curved ....

....the screen. Although (and perhaps because) the final model has flaws resulting from specularities, missing concavities, and imperfect image registration, it unequivocally evokes an uncanny sense of the actual vehicle. 15 Figure 2. 4: Images from a silhouette modeling project by Rick Szeliski [45, 44]. The cup was videotaped on a rotating platform (left) and the extracted contours from this image sequence were used to automatically recover the shape of the cup (right) Although not adequate for general building shapes, silhouette contours could be useful in recovering the approximate shapes ....

Richard Szeliski and Rich Weiss. Robust shape recovery from occluding contours using a linear smoother. Technical Report 93/7, Digital Equipment Corporation, December 1993.

....a single silhouette. Strong prior assumptions are needed to make reconstruction possible in this case (e.g. 10,1,4,9] A second approach is to collect a large number of silhouettes, from known viewpoints, and use them to reconstruct a 3D object using di#erential methods (e.g. 5] 3] 12] [11]) or volume intersection (e.g. 8] 2] These methods can produce accurate approximations to 3D shape, although interestingly, Laurentini[7] shows that exact reconstruction of even very simple polyhedra may require an unbounded number of images. Our current work makes quite di#erent assumptions. ....

Szeliski, R. and Weiss, R., 1998, "Robust Shape Recovery from Occluding Contours Using a Linear Smoother," IJCV 28(1):27-44.

....closest object still not reconstructed is taken into account. This guarantees all the occluders are known before we try to reconstruct an hidden object. We need also to choose the 3D structure we use to represent the objects during the process. One way to do this is to use their contours as in [5, 18, 27, 32], but these methods are very dependent on the contour extraction algorithm that does not give satisfying results with noisy images and real background. We can also optimize a surface so it fits to the image contents using shading information as Fua [10] This approach deals well with untextured ....

Szeliski and Weiss. Robust shape recovery from occluding contours using a linear smoother. Technical Report DECCRL -93-7, Digital Equipment Corporation, Cambridge Research Lab, 93.

....methods as the ones in [35] 14] and [21] cannot deal with situations where profiles are the only available features in the scene. Earlier attempts to solve the problem of reconstruction from image profiles under known motion include [16, 37, 7] and state of the art algorithms can be found in [34, 3, 39]. We use a simple method based on triangulation to reconstruct the model using the estimated motion. Examples using voxel carving [33, 23] are also shown. Details of the 3D reconstruction of the objects are shown in Fig. 14, Fig. 15 and 27 9.6 5 10 15 20 25 5 10 15 20 25 30 35 image index ....

R. Szeliski and R. Weiss. Robust shape recovery from occluding contours using a linear smoothen Int. Journal of Computer Vision, 28(1):27-44, June 1998.

....instance [1, 2, 7] Many of these techniques make interesting, innovative effective use of the duality property between points and planes in homogeneous coordinates. Indeed, this duality is particularly useful to handle 3D object surface reconstruction from occluding contours. But it is only in [10] that optimally combining all the available measurements for better robustness to noise is taken into account and partially solved. Another practical problem also arises from missing data in the occluding contours. On the way to our long term goal to bring surface reconstruction from gray level, ....

R. Szeliski and R. Weiss. Robust shape recovery from occluding contours using a linear smoother. IJCV, 28(1):27-- 44, 1998.

....methods as the ones in [35] 14] and [21] cannot deal with situations where profiles are the only available features in the scene. Earlier attempts to solve the problem of reconstruction from image profiles under known motion include [16, 37, 7] and state of the art algorithms can be found in [34, 3, 39]. We use a simple method based on triangulation to reconstruct the model using the estimated motion. Examples using voxel carving [33, 23] are also shown. Details of the 3D reconstruction of the objects are shown in Fig. 14, Fig. 15 and 27 5 10 15 20 25 30 35 0 5 10 15 Image index 5 10 15 ....

R. Szeliski and R. Weiss. Robust shape recovery from occluding contours using a linear smoother. Int. Journal of Computer Vision, 28(1):27--44, June 1998.

.... spatial and temporal smoothness (Poggio et al. 1985; Bolles et al. 1987; Katayama et al. 1995) or scenes containing curved lines (Bascle and Deriche, 1993) planes (Pritchett and Zisserman, 1998) or texture less surfaces (Cipolla and Blake, 1992; Vaillant and Faugeras, 1992; Laurentini, 1994; Szeliski and Weiss, 1994; Kutulakos and Dyer, 1995) very little is known about scene reconstruction under general conditions. In particular, in the absence of a priori geometric information, what can we infer about the structure of an unknown scene from N arbitrarily positioned cameras at known viewpoints Answering ....

Szeliski, R. and R. Weiss: 1994, `Robust Shape Recovery from Occluding Contours Using a Linear Smoother'. In: C. M. Brown and D. Terzopoulos (eds.): Real-time Computer Vision. Cambridge University Press, pp. 141--165.

....surface without noticeable texture, point and line correspondences may not be easily established. In this case the profile (apparent contour) of the surface is, very often, the only feature available. This calls for the development of a completely different set of techniques, as those found in [32, 7, 4, 30, 22]. This thesis aims at developing simple and practical techniques for the recovery of 1 CHAPTER 1. INTRODUCTION 2 structure and motion from image profiles, making use of the special properties exhibited by the profiles of a smooth curved object performing circular motion. Such techniques will be ....

....lying on the same plane, one of them will be projected onto the plane defined by the others. Cipolla and Blake used the epipolar plane (see figure A.1) as the osculating plane whereas Vaillant and Faugeras used the radial plane spanned by the viewing ray and the surface normal (see figure 2. 2) In [30], Szeliski and Weiss improved the reconstruction by using a linear smoother to compute epipolar curves on the whole surface together with an estimate of uncertainty, but the basic principle is still based on the osculating circle method. ....

R. Szeliski and R. Weiss. Robust shape recovery from occluding contours using a linear smoother. Int. Journal of Computer Vision, 28(1):27--44, June 1998.

....may be required to generate a smooth continuous surface [Hoppe et al. 1992; Szeliski and Tonnesen, 1992] There exist many techniques for extracting 3 D shape from multiple views. For example, we can recover a volumetric description from the binary silhouettes of an object against its background [Szeliski, 1993], compute local optic flow (pixel motion) estimates and convert these into sparse 3 D point estimates [Szeliski, 1991] or track the occluding contours of an object to generate 3 D space curves [Szeliski and Weiss, 1993] A complete survey of 3 D shape extraction techniques is beyond the scope of ....

.... a volumetric description from the binary silhouettes of an object against its background [Szeliski, 1993] compute local optic flow (pixel motion) estimates and convert these into sparse 3 D point estimates [Szeliski, 1991] or track the occluding contours of an object to generate 3 D space curves [Szeliski and Weiss, 1993]. A complete survey of 3 D shape extraction techniques is beyond the scope of this paper. Instead, we present the results of two of our previously developed algorithms applied to an image sequence of a cup rotating on a calibrated turntable (Figure 6a) 7 Full 3 D model recovery 17 (a) b) 175 ....

[Article contains additional citation context not shown here]

R. Szeliski and R. Weiss. Robust shape recovery from occluding contours using a linear smoother. In Image Understanding Workshop, pages 939--948, Morgan Kaufmann Publishers, Washington, D. C., April 1993. A longer version is available as CRL TR 93/7.

....(a) sequence of input panoramic images; b) top view of recovered 3 D points; c) top view of 3 D mesh; and (d) oblique view of texture mapped model. this area. Automatic modeling from images can take the form of reconstruction of an octree model from silhouettes [25] 3 D curves from contours [28] (Figure 1) or simply from 3 D points using panoramic images [10] Figure 2) Examples of 3 D construction systems that require some interactivity include Facade, which uses manual placement of 3 D primitives [5] and Shum et al. s system that makes use of 2 D lines drawn on panoramic mosaics ....

R. Szeliski and R. Weiss. Robust shape recovery from occluding contours using a linear smoother. IJCV, 32(1):27-- 44, June 1998.

No context found.

R. Szeliski, R. Weiss, Robust shape recovery from occluding contours using a linear smoother, Int. J. Comput. Vis. 28 (1) (1998) 27--44.

No context found.

R. Szeliski and R. Weiss. Robust shape recovery from occluding contours using a linear smoother. Int. Journal of Computer Vision, 28(1):27--44, June 1998.

No context found.

R. Szeliski and R. Weiss. Robust shape recovery from occluding contours using a linear smoother. Int. Journal of Computer Vision, 28(1):27--44, June 1998.

No context found.

R. Szeliski and R. Weiss. Robust shape recovery from occluding contours using a linear smoother. Int. Journal of Computer Vision, 28(1):27--44, June 1998.

No context found.

R. Szeliski and R. Weiss. Robust shape recovery from occluding contours using a linear smoother. Int. Journal of Computer Vision, 28(1):27--44, June 1998.

No context found.

R. Szeliski and R. Weiss, "Robust shape recovery from occluding contours using a linear smoother," Int. Journal of Computer Vision, vol. 28, no. 1, pp. 27--44, June 1998.

No context found.

R. Szeliski and R. Weiss. Robust Shape Recovery from Occluding Contours Using a Linear Smoother. Proc. IEEE CVPR, pp. 666-667, 1993.

No context found.

R. Szeliski, R. Weiss, Robust shape recovery from occluding contours using a linear smoother, Int. J. Comput. Vis. 28 (1) (1998) 27--44.

No context found.

R. Szeliski and R. Weiss, "Robust shape recovery from occluding contours using a linear smoother," Digital Equipment Corporation, Cambridge Research Lab, Tech. Rep. DEC-CRL-93-7, 1993.

No context found.

R. Szeliski, R. Weiss, Robust shape recovery from occluding contours using a linear smoother, Int. Journal of Computer Vision 28 (1) (1998) 27--44.

No context found.

R. Szeliski and R. Weiss, "Robust shape recovery from occluding contours using a linear smoother," in Proc. Computer Vision and Pattern Recognition Conf., pp. 666--667, 1993. 19

No context found.

R. Szeliski and R. Weiss, "Robust shape recovery from occluding contours using a linear smoother," in Real-time Computer Vision (C. M. Brown and D. Terzopoulos, eds.), pp. 141--165, Cambridge University Press, 1994.

No context found.

Szeliski, R. and Weiss, R. "Robust Shape Recovery From Occluding Contours Using A Linear Smoother", International Journal of Computer Vision. vol. 28, bo. 1, June 1998, pp. 27-44.

No context found.

Richard Szeliski and Richard Weiss. Robust shape recovery from occluding contours using a linear smoother. Intl Journal of Computer Vision, 28(1):1998, 1998.

No context found.

Szeliski, R. and Weiss, R. (1993). Robust shape recovery from occluding contours using a linear smoother. Technical report, Cambridge Research Laboratory, Digital Equipment Corporation.

First 50 documents

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