C. L. Zitnick and J. A. Webb, "Multi-baseline stereo using surface extraction," Tech. Rep. CMU-CS-96-196, Carnegie Mellon University, Pittsburgh, PA, November 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....flash projects a pattern of vertical stripes on the subject. At the same time, six b w digital cameras photograph the illuminated area from different angles. An sample stripe image is shown in Figure 3(c) An additional digital color camera provides a texture image. A multiview stereo algorithm [5], part of the software suite provided with the Virtuoso system, computes a triangle mesh approximating the scanned area. In our scanning conditions each scan typically covered a 20 cm by 20 cm area and comprised on average about 10,000 measured points. The typical intersample distance for these ....

C. Lawrence Zitnick and Jon A. Webb, "Multi-baseline stereo using surface extraction," Tech. Rep. CMU-CS-96-196, School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213-3890, Nov. 1996.

....results, more research is needed to assess the applicability of these methods for noisier stereo derived models. A disadvantage of the method is that the original images are not used in the merging process so it is difficult to assess the photo integrity of the reconstructions. Zitnick and Webb [50] described a scene space stereo technique that detects and reconstructs scene regions that appear unoccluded in a set of input images. They noted that the correspondence problem is ill posed in the presence of occlusion, but can be solved when occlusion does not occur. By posing the problem as one ....

C. Lawrence Zitnick and Jon A. Webb. Multi-baseline stereo using surface extraction. Technical Report CMU-CS-96196, Carnegie Mellon University, Pittsburgh, PA, November 1996. 32

....features in the images. Correspondences are identified by modeling the statistical likelihood of accidental accumulation and thresholding the votes to achieve a desired false positive rate. As with [SD95a] occlusions are not explicitly modeled in 65 the Space Sweep approach. Zitnick and Webb [ZW96] described a scene space stereo technique that reconstructs scene regions that project unoccluded to the basis images. They noted that the correspondence problem is fundamentally ill posed in the presence of occlusion, but can be solved when occlusion does not occur. By posing the problem as one ....

C. Lawrence Zitnick and Jon A. Webb. Multi-baseline stereo using surface extraction. Technical Report CMU-CS-96-196, Carnegie Mellon University, Pittsburgh, PA, November 1996.

....the largest potential surface is correct. Intuitively this makes sense with the four post picket fence. However, the largest surface is not always the correct surface. As shown in figure 4(b) it is possible for a false surface to be the largest surface when it is occluded. As discussed in [23], this case is rare and its occurrence can be further minimized if smaller baselines are used. Figure 5 demonstrates how well the algorithm can deal with the locally ambiguous stereo matching problem with repetitive texture. The synthetic example scene is a slanted roof, figure 5(a) with a ....

C. L. Zitnick and J. A. Webb, "Multi-baseline stereo using surface extraction," CMU Technical Report CMU-CS-96-196, 1996.

No context found.

C. L. Zitnick and J. A. Webb, "Multi-baseline stereo using surface extraction," Tech. Rep. CMU-CS-96-196, Carnegie Mellon University, Pittsburgh, PA, November 1996.

No context found.

C. L. Zitnick and J. A. Webb, "Multi-baseline stereo using surface extraction," Tech. Rep. CMU-CS-96-196, Carnegie Mellon University, Pittsburgh, PA, November 1996.

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