Bayesian Shape from Silhouettes

详细信息    查看全文

• 作者：Donghoon Kim (18)
  Rozenn Dahyot (18)
  
• 关键词：shape from silhouettes ; kernel density estimates ; Mean ; shift algorithm ; Gaussian stack ; KNN ; visual hull
• 刊名：Lecture Notes in Computer Science
• 出版年：2012
• 出版时间：2012
• 年：2012
• 卷：7252
• 期：1
• 页码：90-101
• 全文大小：832KB
• 参考文献：1. Kim, D., Ruttle, J., Dahyot, R.: 3d shape estimation from silhouettes using mean-shift. In: IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), pp. 1430鈥?433 (2010)
  2. Dyer, C.R.: Volumetric scene reconstruction from multiple views. In: Davis, L.S. (ed.) Foundation of Image Understanding, pp. 469鈥?89. Kluwer, Boston (2001) CrossRef
  3. Franco, J.S., Boyer, E.: Efficient polyhedral modeling from silhouettes. IEEE Transactions on Pattern Analysis and Machine Intelligence, 414鈥?27 (2008)
  4. Zimmermann, K., Svoboda, T., Matas, J.: Multiview 3d tracking with an incrementally constructed 3d model. In: The Third International Symposium on 3D Data Processing, Visualization, and Transmission (2006)
  5. Laurentini, A.: How far 3d shapes can be understood from 2d silhouettes. IEEE Transactions on Pattern Analysis and Machine Intelligence聽17(2), 188鈥?95 (1995) CrossRef
  6. Martin, W.N., Aggarwal, J.K.: Volumetric description of objects from multiple views. IEEE Transactions on Pattern Analysis and Machine Intelligence聽5(2), 150鈥?58 (1987) CrossRef
  7. Potmesil, M.: Generating octree models of 3d objects from their silhouettes in a sequence of images. Computer Vision, Graphics, and Image Processing聽40, 1鈥?9 (1987) CrossRef
  8. Srivasan, P., Liang, P., Hackwood, S.: Computational geometric methods in volumetric intersections for 3d reconstruction. Pattern Recognition聽23(8), 843鈥?57 (1990) CrossRef
  9. Baumgart, B.G.: Geometric modeling for computer vision. Technical report, Artificial Intelligence Laboratory, Stanford University (1974)
  10. Lazebnik, S., Boyer, E., Ponce, J.: On computing exact visual hulls of solids bounded by smooth surfaces. In: IEEE Conference on Computer Vision and Pattern Recognition, vol.聽1, pp. 156鈥?61 (2001)
  11. Matusik, W., Buehler, C., McMillan, L.: Polyhedral visual hulls for real-time rendering. In: Eurographics Workshop on Rendering, pp. 115鈥?26 (2001)
  12. Sullivan, S., Ponce, J.: Automatic model construction, pose estimation, and object recognition from photographs using triangular splines. IEEE Transactions on Pattern Analysis and Machine Intelligence聽20, 1091鈥?096 (1998) CrossRef
  13. Ruttle, J., Manzke, M., Dahyot, R.: Smooth kernel density estimate for multiple view reconstruction. In: The 7th European Conference for Visual Media Production (2010)
  14. Exner, D., Bruns, E., Kurz, D., Grundh枚fer, A., Bimber, O.: Fast and robust camshift tracking. In: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, pp. 9鈥?6 (2010)
  15. Srinivasan, B.V., Hu, Q., Duraiswami, R.: Graphical processors for speeding up kernel machines. In: Workshop on High Performance Analytics - Algorithms, Implementations, and Applications, Siam Conference on Data Mining (2010)
  16. Cootes, T.F., Edwards, G.J., Taylor, C.: Active appearance models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 681鈥?85 (2001)
  17. Lefebvre, S., Hoppe, H.: Parallel controllable texture synthesis. In: ACM SIGGRAPH (2005)
  18. Geusebroek, J.M., Burghouts, G.J., Smeulders, A.W.M.: The amsterdam library of object images. International Journal Computer Vision聽61(1), 103鈥?12 (2005) CrossRef
  19. Kim, D.: 3D Object Reconstruction using Multiple Views. PhD thesis, School of Computer Science and Statistics, Trinity College Dublin, Dublin, Ireland (2011)
  20. Paysan, P., Knothe, R., Amberg, B., Romdhani, S., Vetter, T.: A 3d face model for pose and illumination invariant face recognition. In: The 6th IEEE International Conference on Advanced Video and Signal based Surveillance (AVSS) for Security, Safety and Monitoring in Smart Environments (2009)
  
• 作者单位：Donghoon Kim (18)
  Rozenn Dahyot (18)
  
  18. School of Computer Science and Statistics, Trinity College Dublin, Ireland
  

文摘

This paper extends the likelihood kernel density estimate of the visual hull proposed by Kim et al [1] by introducing a prior. Inference of the shape is performed using a meanshift algorithm over a posterior kernel density function that is refined iteratively using both a multiresolution framework (to avoid local maxima) and using KNN for selecting the best reconstruction basis at each iteration. This approach allows us to recover concave areas of the shape that are usually lost when estimating the visual hull.