Descripción y análisis de forma basado en la invarianza a isometrías de los esqueletos topológicos

dc.contributor.advisorBranch Bedoya, John William
dc.contributor.authorPatiño Cortés, Diego Alberto
dc.contributor.researchgroupGIDIA - Grupo de Investigación en Inteligencia Artificialspa
dc.date.accessioned2021-06-19T13:52:39Z
dc.date.available2021-06-19T13:52:39Z
dc.date.issued2019-06
dc.descriptionilustracionesspa
dc.description.abstractIn this dissertation, we explore the problem of how to describe the shape of an object in 2D and 3D with a set of features that are invariant to isometric transformations. We focus to based our approach on the well-known Medial Axis Transform and its topological properties. We aim to study two problems. The first is how to find a shape representation of a segmented object that exhibits rotation, translation, and reflection invariance. The second problem is how to build a machine learning pipeline that uses the isometric invariance of the shape representation to do both classification and retrieval. Our proposed solution demonstrates competitive results compared to state-of-the-art approaches. We based our shape representation on the medial axis transform (MAT), sometimes called the topological skeleton. Accepted and well-studied properties of the medial axis include: homotopy preservation, rotation invariance, mediality, one pixel thickness, and the ability to fully reconstruct the object. These properties make the MAT a suitable input to create shape features; however, several problems arise because not all skeletonization methods satisfy all the above-mentioned properties at the same time. In general, skeletons based on thinning approaches preserve topology but are noise sensitive and do not allow a proper reconstruction. They are also not invariant to rotations. Voronoi skeletons also preserve topology and are rotation invariant, but do not have information about the thickness of the object, making reconstruction impossible. The Voronoi skeleton is an approximation of the real skeleton. The denser the sampling of the boundary, the better the approximation; however, a denser sampling makes the Voronoi diagram more computationally expensive. In contrast, distance transform methods allow the reconstruction of the original object by providing the distance from every pixel in the skeleton to the boundary. Moreover, they exhibit an acceptable degree of the properties listed above, but noise sensitivity remains an issue. Therefore, we selected distance transform medial axis methods as our skeletonization strategy, and focused on creating a new noise-free approach to solve the contour noise problem. To effectively classify an object, or perform any other task with features based on its shape, the descriptor needs to be a normalized, compact form: $\Phi$ should map every shape $\Omega$ to the same vector space $\mathrm{R}^{n}$. This is not possible with skeletonization methods because the skeletons of different objects have different numbers of branches and different numbers of points, even when they belong to the same category. Consequently, we developed a strategy to extract features from the skeleton through the map $\Phi$, which we used as an input to a machine learning approach. After developing our method for robust skeletonization, the next step is to use such skeleton into the machine learning pipeline to classify object into previously defined categories. We developed a set of skeletal features that were used as input data to the machine learning architectures. We ran experiments on MPEG7 and ModelNet40 dataset to test our approach in both 2D and 3D. Our experiments show results comparable with the state-of-the-art in shape classification and retrieval. Our experiments also show that our pipeline and our skeletal features exhibit some degree of invariance to isometric transformations. In this study, we sought to design an isometric invariant shape descriptor through robust skeletonization enforced by a feature extraction pipeline that exploits such invariance through a machine learning methodology. We conducted a set of classification and retrieval experiments over well-known benchmarks to validate our proposed method. (Tomado de la fuente)eng
dc.description.abstractEn esta disertación se explora el problema de cómo describir la forma de un objeto en 2D y 3D con un conjunto de características que sean invariantes a transformaciones isométricas. La metodología propuesta en este documento se enfoca en la Transformada del Eje Medio (Medial Axis Transform) y sus propiedades topológicas. Nuestro objetivo es estudiar dos problemas. El primero es encontrar una representación matemática de la forma de un objeto que exhiba invarianza a las operaciones de rotación, translación y reflexión. El segundo problema es como construir un modelo de machine learning que use esas invarianzas para las tareas de clasificación y consulta de objetos a través de su forma. El método propuesto en esta tesis muestra resultados competitivos en comparación con otros métodos del estado del arte. En este trabajo basamos nuestra representación de forma en la transformada del eje medio, a veces llamada esqueleto topológico. Algunas propiedades conocidas y bien estudiadas de la transformada del eje medio son: conservación de la homotopía, invarianza a la rotación, su grosor consiste en un solo pixel (1D), y la habilidad para reconstruir el objeto original a través de ella. Estas propiedades hacen de la transformada del eje medio un punto de partida adecuado para crear características de forma. Sin embargo, en este punto surgen varios problemas dado que no todos los métodos de esqueletización satisfacen, al mismo tiempo, todas las propiedades mencionadas anteriormente. En general, los esqueletos basados en enfoques de erosión morfológica conservan la topología del objeto, pero son sensibles al ruido y no permiten una reconstrucción adecuada. Además, no son invariantes a las rotaciones. Otro método de esqueletización son los esqueletos de Voronoi. Los esqueletos de Voronoi también conservan la topología y son invariantes a la rotación, pero no tienen información sobre el grosor del objeto, lo que hace imposible su reconstrucción. Cuanto más denso sea el muestreo del contorno del objeto, mejor será la aproximación. Sin embargo, un muestreo más denso hace que el diagrama de Voronoi sea más costoso computacionalmente. Por el contrario, los métodos basados en la transformada de la distancia permiten la reconstrucción del objeto original, ya que proporcionan la distancia desde cada píxel del esqueleto hasta su punto más cercano en el contorno. Además, exhiben un grado aceptable de las propiedades enumeradas anteriormente, aunque la sensibilidad al ruido sigue siendo un problema. Por lo tanto, en este documento seleccionamos los métodos basados en la transformada de la distancia como nuestra estrategia de esqueletización, y nos enfocamos en crear un nuevo enfoque que resuelva el problema del ruido en el contorno. Para clasificar eficazmente un objeto o realizar cualquier otra tarea con características basadas en su forma, el descriptor debe ser compacto y estar normalizado: $\Phi$ debe relacionar cada forma $\Omega$ al mismo espacio vectorial $\mathrm{R}^{n}$. Esto no es posible con los métodos de esqueletización en el estado del arte, porque los esqueletos de diferentes objetos tienen diferentes números de ramas y diferentes números de puntos incluso cuando pertenecen a la misma categoría. Consecuentemente, en nuestra propuesta desarrollamos una estrategia para extraer características del esqueleto a través de la función $\Phi$, que usamos como entrada para un enfoque de aprendizaje automático. % TODO completar con resultados. Después de desarrollar nuestro método de esqueletización robusta, el siguiente paso es usar dicho esqueleto en un modelo de aprendizaje de máquina para clasificar el objeto en categorías previamente definidas. Para ello se desarrolló un conjunto de características basadas en el eje medio que se utilizaron como datos de entrada para la arquitectura de aprendizaje automático. Realizamos experimentos en los conjuntos de datos: MPEG7 y ModelNet40 para probar nuestro enfoque tanto en 2D como en 3D. Nuestros experimentos muestran resultados comparables con el estado del arte en clasificación y consulta de formas (retrieval). Nuestros experimentos también muestran que el modelo desarrollado junto con nuestras características basadas en el eje medio son invariantes a las transformaciones isométricas. (Tomado de la fuente)spa
dc.description.degreelevelDoctoradospa
dc.description.degreenameDoctor en Ingenieríaspa
dc.description.researchareaVisión por computadora y aprendizaje automáticospa
dc.description.sponsorshipBeca para Doctorados Nacionales de Colciencias, convocatoria 725 de 2015spa
dc.format.extent104 páginasspa
dc.format.mimetypeapplication/pdfspa
dc.identifier.instnameUniversidad Nacional de Colombiaspa
dc.identifier.reponameRepositorio Institucional Universidad Nacional de Colombiaspa
dc.identifier.repourlhttps://repositorio.unal.edu.co/spa
dc.identifier.urihttps://repositorio.unal.edu.co/handle/unal/79654
dc.language.isoengspa
dc.publisherUniversidad Nacional de Colombiaspa
dc.publisher.branchUniversidad Nacional de Colombia - Sede Medellínspa
dc.publisher.departmentDepartamento de la Computación y la Decisiónspa
dc.publisher.facultyFacultad de Minasspa
dc.publisher.placeMedellínspa
dc.publisher.programMedellín - Minas - Doctorado en Ingeniería - Sistemasspa
dc.relation.referencesAtabay, H. A. (2017). Binary shape classification using convolutional neural networks. IIOAB Journal, 7(October 2016):332–336.spa
dc.relation.referencesAtienza, R. (2019). Pyramid u-network for skeleton extraction from shape points. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops.spa
dc.relation.referencesAlajlan, N., Kamel, M., and Freeman, G. (2008). Geometry-based image retrieval in binary image databases. IEEE Transactions on Pattern Analysis and Machine Intelligence, 30(6):1003–1013.spa
dc.relation.referencesAtienza, R. (2019). Pyramid u-network for skeleton extraction from shape points. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops.spa
dc.relation.referencesAttalla, E. and Siy, P. (2005). Robust shape similarity retrieval based on contour segmentation polygonal multiresolution and elastic matching. Pattern Recognition, 38(12):2229–2241.spa
dc.relation.referencesAu, O. K.-C., Tai, C.-L., Chu, H.-K., Cohen-Or, D., and Lee, T.-Y. (2008). Skeleton extraction by mesh contraction. ACM Trans. Graph., 27(3):44:1–44:10.spa
dc.relation.referencesAubert, G. and Aujol, J. F. (2014). Poisson skeleton revisited: A new mathematical perspective. Journal of Mathematical Imaging and Vision, 48(1):149–159.spa
dc.relation.referencesAubry, M., Schlickewei, U., and Cremers, D. (2011). The wave kernel signature: A quantum mechanical approach to shape analysis. In 2011 IEEE International Conference on Computer Vision Workshops (ICCV Workshops). IEEE.spa
dc.relation.referencesBai, X., Liu, W., Tu, Z., and Angeles, L. (2009). Integrating Contour and Skeleton for Shape Classification. In Workshop on NORDIA (in conjunction with ICCV09).spa
dc.relation.referencesBai, X., Yang, X., Latecki, L., Liu, W., and Tu, Z. (2010). Learning context-sensitive shape similarity by graph transduction. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32(5):861–874.spa
dc.relation.referencesBelkin, M., Niyogi, P., and Sindhwani, V. (2006). Manifold regularization: A geometric framework for learning from labeled and unlabeled examples. J. Mach. Learn. Res., 7:2399–2434.spa
dc.relation.referencesBelongie, S., Malik, J., and Puzicha, J. (2002). Shape matching and object recognition using shape contexts. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(4):509–522.spa
dc.relation.referencesBeristain, A. and Grana, M. (2010). Pruning algorithm for voronoi skeletons. Electronics Letters, 46(1):39– 41.spa
dc.relation.referencesBernard, T. M. and Manzanera, A. (1999). Improved low complexity fully parallel thinning algorithm. In Proceedings 10th International Conference on Image Analysis and Processing, pages 215–220.spa
dc.relation.referencesBhuptani, N. and Talati, B. (2014). Variations in shape context descriptor: A survey. International Journal of Computer Applications, 90(12):29–33.spa
dc.relation.referencesBiasotti, S., Falcidieno, B., Giorgi, D., and Spagnuolo, M. (2014). Mathematical Tools for Shape Analysis and Description. Synthesis Lectures on Computer Graphics and Animation. Morgan & Claypool Publishers.spa
dc.relation.referencesBlum, H. (1967). A Transformation for Extracting New Descriptors of Shape. In Wathen-Dunn, W., editor, Models for the Perception of Speech and Visual Form, pages 362–380. MIT Press, Cambridge.spa
dc.relation.referencesButt, M. A. and Maragos, P. (1998). Optimum design of chamfer distance transforms. IEEE Transactions on Image Processing, 7(10):1477–1484.spa
dc.relation.referencesChang, A. X., Funkhouser, T., Guibas, L., Hanrahan, P., Huang, Q., Li, Z., Savarese, S., Savva, M., Song, S., Su, H., Xiao, J., Yi, L., and Yu, F. (2015). ShapeNet: An Information-Rich 3D Model Reposi- tory. Technical Report arXiv:1512.03012 [cs.GR], Stanford University — Princeton University — Toyota Technological Institute at Chicago.spa
dc.relation.referencesChaudhari, A. J., Leahy, R. M., Wise, B. L., Lane, N. E., Badawi, R. D., and Joshi, A. A. (2014). Global point signature for shape analysis of carpal bones. Physics in Medicine and Biology, 59(4):961–973.spa
dc.relation.referencesChaudhry, R., Ofli, F., Kurillo, G., Bajcsy, R., and Vidal, R. (2013). Bio-inspired dynamic 3d discriminative skeletal features for human action recognition. In 2013 IEEE Conference on Computer Vision and Pattern Recognition Workshops. IEEE.spa
dc.relation.referencesChaussard, J., Couprie, M., and Talbot, H. (2011). Robust skeletonization using the discrete λ-medial axis. Pattern Recognition Letters, 32(9):1384–1394.spa
dc.relation.referencesChui, C. K., Lin, S.-B., and Zhou, D.-X. (2019). Deep neural networks for rotation-invariance approximation and learning. ArXiv, abs/1904.01814.spa
dc.relation.referencesCohen, T. S., Geiger, M., K¨ohler, J., and Welling, M. (2018). Spherical CNNs. In International Conference on Learning Representations.spa
dc.relation.referencesCouprie, M., Coeurjolly, D., and Zrour, R. (2007). Discrete bisector function and Euclidean skeleton in 2D and 3D. Image and Vision Computing, 25(10):1543–1556.spa
dc.relation.referencesdo Carmo, M. (1992). Riemannian Geometry. Mathematics (Boston, Mass.). Birkh¨auser.spa
dc.relation.referencesDrew, M. S., Lee, T. K., and Rova, A. (2009). Shape retrieval with eigen-CSS search. Image and Vision Computing, 27(6):748–755.spa
dc.relation.referencesDubuisson, M.-P. and Jain, A. (2002). A modified Hausdorff distance for object matching. In Proceedings of 12th International Conference on Pattern Recognition, volume 1, pages 566–568. IEEE Comput. Soc. Press.spa
dc.relation.referencesEsteves, C., Allen-Blanchette, C., Makadia, A., and Daniilidis, K. (2018a). Learning so(3) equivariant representations with spherical cnns. In Ferrari, V., Hebert, M., Sminchisescu, C., and Weiss, Y., editors, Computer Vision – ECCV 2018, pages 54–70, Cham. Springer International Publishing.spa
dc.relation.referencesEsteves, C., Allen-Blanchette, C., Zhou, X., and Daniilidis, K. (2018b). Polar transformer networks. In 6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, April 30 - May 3, 2018, Conference Track Proceedings.spa
dc.relation.referencesFelzenszwalb, P. F. and Schwartz, J. D. (2007). Hierarchical matching of deformable shapes. In 2007 IEEE Conference on Computer Vision and Pattern Recognition. IEEE.spa
dc.relation.referencesFeng, Y., Zhang, Z., Zhao, X., Ji, R., and Gao, Y. (2018). Gvcnn: Group-view convolutional neural networks for 3d shape recognition. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR).spa
dc.relation.referencesFigueiredo, M. A. T., Leitao, J. M. N., and Jain, A. K. (2000). Unsupervised contour representation and estimation using b-splines and a minimum description length criterion. IEEE Transactions on Image Processing, 9(6):1075–1087.spa
dc.relation.referencesFreifeld, O. and Black, M. J. (2012). Lie Bodies: A Manifold Representation of 3D Human Shape. In Leibe, B., Matas, J., Sebe, N., and Welling, M., editors, European Conference on Computer Vision, number October 2012 in Lecture Notes in Computer Science, pages 1–14. Springer International Publishing, Cham.spa
dc.relation.referencesGao, F., Wei, G., Xin, S., Gao, S., and Zhou, Y. (2018). 2D skeleton extraction based on heat equation. Computers and Graphics (Pergamon), 74:99–108.spa
dc.relation.referencesGao, Z., Yu, Z., and Pang, X. (2014). A compact shape descriptor for triangular surface meshes. Computer- Aided Design, 53:62–69.spa
dc.relation.referencesGiesen, J., Miklos, B., Pauly, M., and Wormser, C. (2009). The scale axis transform. In Proceedings of the 25th annual symposium on Computational geometry - SCG ’09, page 106, New York, New York, USA. ACM Press.spa
dc.relation.referencesGorelick, L., Galun, M., and Sharon, E. (2006). Shape Representation and Classification Using the Poisson Equation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 28(12):1991–2005.spa
dc.relation.referencesGrigorescu, C. and Petkov, N. (2003). Distance sets for shape filters and shape recognition. IEEE Transac- tions on Image Processing, 12(10):1274–1286.spa
dc.relation.referencesHesselink, W. H. and Roerdink, J. B. (2008). Euclidean skeletons of digital image and volume data in linear time by the integer medial axis transform. IEEE Transactions on Pattern Analysis and Machine Intelligence, 30(12):2204–2217.spa
dc.relation.referencesHuang, J. and You, S. (2016). Point cloud labeling using 3d convolutional neural network. In 2016 23rd International Conference on Pattern Recognition (ICPR). IEEE.spa
dc.relation.referencesKanezaki, A. (2016). Rotationnet: Learning object classification using unsupervised viewpoint estimation. CoRR, abs/1603.06208.spa
dc.relation.referencesKendall, D. G. (1977). The diffusion of shape. Advances in Applied Probability, 9(3):428–430.spa
dc.relation.referencesKendall, D. G. (1984a). Shape manifolds, procrustean metrics, and complex projective spaces. Bulletin of the London Mathematical Society, 16(2):81–121.spa
dc.relation.referencesKendall, D. G. (1984b). Shape manifolds, procrustean metrics, and complex projective spaces. Bulletin of the London Mathematical Society, 16(2):81–121.spa
dc.relation.referencesKhotanzad, A. and Hong, Y. H. (1990). Invariant image recognition by zernike moments. IEEE Transactions on Pattern Analysis and Machine Intelligence, 12(5):489–497.spa
dc.relation.referencesKim, H.-K. and Kim, J.-D. (2000). Region-based shape descriptor invariant to rotation, scale and translation. Signal Processing: Image Communication, 16(1-2):87–93.spa
dc.relation.referencesKING, D. B., WERTHEIMER, M., KELLER, H., and CROCHETIE`RE, K. (1994). The legacy of max wertheimer and gestalt psychology. Social Research, 61(4):907–935.spa
dc.relation.referencesKoffka, K. (1999). Principles of Gestalt Psychology. Cognitive psychology]. Routledge.spa
dc.relation.referencesKokkinos, I., Bronstein, M. M., Litman, R., and Bronstein, A. M. (2012). Intrinsic shape context descriptors for deformable shapes. In 2012 IEEE Conference on Computer Vision and Pattern Recognition, pages 159–166.spa
dc.relation.referencesKondor, R. and Trivedi, S. (2018). On the generalization of equivariance and convolution in neural networks to the action of compact groups. arXiv preprint arXiv:1802.03690.spa
dc.relation.referencesKrizhevsky, A., Sutskever, I., and Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. In Pereira, F., Burges, C. J. C., Bottou, L., and Weinberger, K. Q., editors, Advances in Neural Information Processing Systems 25, pages 1097–1105. Curran Associates, Inc.spa
dc.relation.referencesKulon, D., Wang, H., Gu¨ler, R. A., Bronstein, M., and Zafeiriou, S. P. (2019). Single image 3d hand reconstruction with mesh convolutions. ArXiv, abs/1905.01326.spa
dc.relation.referencesLaga, H. (2018). A survey on nonrigid 3d shape analysis. In Academic Press Library in Signal Processing, Volume 6, pages 261–304. Elsevier.spa
dc.relation.referencesLatecki, L. J. and Lakamper, R. (2000). Shape similarity measure based on correspondence of visual parts. IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(10):1185–1190.spa
dc.relation.referencesLecun, Y., Bottou, L., Bengio, Y., and Haffner, P. (1998). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324.spa
dc.relation.referencesLee, R. N. (1984). Two-dimensional critical point configuration graphs. IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-6(4):442–450.spa
dc.relation.referencesLi, C. and Ben Hamza, A. (2013). A multiresolution descriptor for deformable 3d shape retrieval. Vis. Comput., 29(6-8):513–524.spa
dc.relation.referencesLi, H., Sun, L., Wu, X., and Cai, Q. (2018). Scale-invariant wave kernel signature for non-rigid 3d shape retrieval. In 2018 IEEE International Conference on Big Data and Smart Computing (BigComp). IEEE.spa
dc.relation.referencesLi, M., Chen, S., Chen, X., Zhang, Y., Wang, Y., and Tian, Q. (2019). Actional-structural graph convolu- tional networks for skeleton-based action recognition. ArXiv, abs/1904.12659.spa
dc.relation.referencesLi, R., Bu, G., and Wang, P. (2017). An automatic tree skeleton extracting method based on point cloud of terrestrial laser scanner. International Journal of Optics, 2017:1–11.spa
dc.relation.referencesLimberger, F. A. and Wilson, R. C. (2015). Feature encoding of spectral signatures for 3d non-rigid shape retrieval. In Procedings of the British Machine Vision Conference 2015. British Machine Vision Associa- tion.spa
dc.relation.referencesLing, H., Member, S., and Jacobs, D. W. (2007). Shape Classification Using the Inner-Distance. IEEE Transactions on Pattern Analysis and Machine Intelligence, 29(2):286–299.spa
dc.relation.referencesLitany, O., Bronstein, A. M., Bronstein, M. M., and Makadia, A. (2018). Deformable shape completion with graph convolutional autoencoders. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1886–1895.spa
dc.relation.referencesLitman, R. and Bronstein, A. M. (2014). Learning spectral descriptors for deformable shape correspondence. IEEE Transactions on Pattern Analysis and Machine Intelligence, 36(1):171–180.spa
dc.relation.referencesLiu, Y. K. and Zˇalik, B. (2005). An efficient chain code with huffman coding. Pattern Recognition, 38(4):553– 557.spa
dc.relation.referencesLowe, D. G. (2004). Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision, 60(2):91–110.spa
dc.relation.referencesMarie, R., Labbani-Igbida, O., and Mouaddib, E. M. (2016). The Delta Medial Axis: A fast and robust algorithm for filtered skeleton extraction. Pattern Recognition, 56:26–39.spa
dc.relation.referencesMasoumi, M., Li, C., and Hamza, A. B. (2016). A spectral graph wavelet approach for nonrigid 3d shape retrieval. Pattern Recognition Letters, 83:339–348.spa
dc.relation.referencesMaturana, D. and Scherer, S. (2015). Voxnet: A 3d convolutional neural network for real-time object recognition. In 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 922–928.spa
dc.relation.referencesMcNeill, G. and Vijayakumar, S. (2005). 2d shape classification and retrieval. In Proceedings of the 19th International Joint Conference on Artificial Intelligence, IJCAI’05, pages 1483–1488, San Francisco, CA, USA. Morgan Kaufmann Publishers Inc.spa
dc.relation.referencesMcNeill, G. and Vijayakumar, S. (2006). Hierarchical procrustes matching for shape retrieval. In Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1 (CVPR’06). IEEE.spa
dc.relation.referencesMiklos, B., Giesen, J., and Pauly, M. (2010). Discrete scale axis representations for 3d geometry. ACM Trans. Graph., 29(4):101:1–101:10.spa
dc.relation.referencesMokhtarian, F., Abbasi, S., and Kittler, J. (1998). Efficient and robust retrieval by shape content through cur- vature scale space. In Series on Software Engineering and Knowledge Engineering, pages 51–58. WORLD SCIENTIFIC.spa
dc.relation.referencesMokhtarian, F. and Bober, M. (2003). Curvature Scale Space Representation: Theory, Applications, and MPEG-7 Standardization. Springer Netherlands.spa
dc.relation.referencesMokhtarian, F. and Mackworth, A. (1992). A theory of multiscale, curvature-based shape representation for planar curves. IEEE Transactions on Pattern Analysis and Machine Intelligence, 14(8):789–805. cited By 722.spa
dc.relation.referencesMori, G., Belongie, S., and Malik, J. (2005). Efficient shape matching using shape contexts. IEEE Transac- tions on Pattern Analysis and Machine Intelligence, 27(11):1832–1837.spa
dc.relation.referencesNava-Yazdani, E., Hege, H.-C., Sullivan, T., and von Tycowicz, C. (2019). Geodesic analysis in kendall’s shape space with epidemiological applications. arXiv preprint arXiv:1906.11950.spa
dc.relation.referencesOgniewicz, R. and Ilg, M. (1992). Voronoi skeletons: theory and applications. In Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pages 63–69.spa
dc.relation.referencesPeter, A., Rangarajan, A., and Ho, J. (2008). Shape l’anerouge: Sliding wavelets for indexing and retrieval. In 2008 IEEE Conference on Computer Vision and Pattern Recognition. IEEE.spa
dc.relation.referencesPeters, R. and Ledoux, H. (2016). Robust approximation of the medial axis transform of LiDAR point clouds as a tool for visualisation. Computers & Geosciences, 90:123–133.spa
dc.relation.referencesPickup, D., Sun, X., Rosin, P. L., and Martin, R. R. (2016a). Skeleton-based canonical forms for non-rigid 3d shape retrieval. Computational Visual Media, 2(3):231–243.spa
dc.relation.referencesPickup, D., Sun, X., Rosin, P. L., Martin, R. R., Cheng, Z., Lian, Z., Aono, M., Hamza, A. B., Bronstein, A., Bronstein, M., Bu, S., Castellani, U., Cheng, S., Garro, V., Giachetti, A., Godil, A., Isaia, L., Han, J., Johan, H., Lai, L., Li, B., Li, C., Li, H., Litman, R., Liu, X., Liu, Z., Lu, Y., Sun, L., Tam, G., Tatsuma, A., and Ye, J. (2016b). Shape retrieval of non-rigid 3d human models. International Journal of Computer Vision, 120(2):169–193.spa
dc.relation.referencesPostolski, M., Couprie, M., and Janaszewski, M. (2014). Scale filtered Euclidean medial axis and its hierarchy. Computer Vision and Image Understanding, 129:89–102.spa
dc.relation.referencesPumarola, A., Agudo, A., Porzi, L., Sanfeliu, A., Lepetit, V., and Moreno-Noguer, F. (2018). Geometry- aware network for non-rigid shape prediction from a single view. 2018 IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, pages 4681–4690.spa
dc.relation.referencesPunam K. Saha, G. B. and de Baja (Eds.), G. S. (2017). Skeletonization. Theory, Methods and Applications. Elsevier Science, 1st edition edition.spa
dc.relation.referencesQi, C. R., Su, H., Mo, K., and Guibas, L. J. (2017a). Pointnet: Deep learning on point sets for 3d classification and segmentation. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 77–85.spa
dc.relation.referencesQi, C. R., Su, H., Niessner, M., Dai, A., Yan, M., and Guibas, L. J. (2016). Volumetric and multi-view cnns for object classification on 3d data. arXiv preprint arXiv:1604.03265.spa
dc.relation.referencesQi, C. R., Yi, L., Su, H., and Guibas, L. J. (2017b). Pointnet++: Deep hierarchical feature learning on point sets in a metric space. arXiv preprint arXiv:1706.02413.spa
dc.relation.referencesQiu, T., Yan, Y., and Lu, G. (2011). A medial axis extraction algorithm for the processing of combustion flame images. In 2011 Sixth International Conference on Image and Graphics, pages 182–186.spa
dc.relation.referencesReuter, M., Wolter, F.-E., and Peinecke, N. (2006). Laplace–beltrami spectra as ‘shape-DNA’ of surfaces and solids. Computer-Aided Design, 38(4):342–366.spa
dc.relation.referencesRumpf, M. and Preusser, T. (2002). A level set method for anisotropic geometric diffusion in 3d image processing. SIAM Journal on Applied Mathematics, 62(5):1772–1793.spa
dc.relation.referencesRustamov, R. M. (2007). Laplace-beltrami eigenfunctions for deformation invariant shape representation. In Proceedings of the Fifth Eurographics Symposium on Geometry Processing, SGP ’07, pages 225–233, Aire-la-Ville, Switzerland, Switzerland. Eurographics Association.spa
dc.relation.referencesSafar, M. H. and Shahabi, C. (2003). Shape Analysis and Retrieval of Multimedia Objects. Springer US.spa
dc.relation.referencesSaha, P. K., Borgefors, G., and Sanniti di Baja, G. (2016). A survey on skeletonization algorithms and their applications. Pattern Recognition Letters, 76:3–12.spa
dc.relation.referencesSato, M., Bitter, I., Bender, M. A., Kaufman, A. E., and Nakajima, M. (2000). Teasar: tree-structure extraction algorithm for accurate and robust skeletons. In Proceedings the Eighth Pacific Conference on Computer Graphics and Applications, pages 281–449.spa
dc.relation.referencesSebastian, T., Klein, P., and Kimia, B. (2003). On aligning curves. IEEE Transactions on Pattern Analysis and Machine Intelligence, 25(1):116–125.spa
dc.relation.referencesSebastian, T. B., Klein, P. N., and Kimia, B. B. (2004). Recognition of shapes by editing shock graphs. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 00(528):755–762.spa
dc.relation.referencesShen, W., Wang, X., Yao, C., and Bai, X. (2014a). Shape recognition by combining contour and skeleton into a mid-level representation. In CCPR.spa
dc.relation.referencesShen, W., Wang, X., Yao, C., and Bai, X. (2014b). Shape recognition by combining contour and skeleton into a mid-level representation. In Communications in Computer and Information Science, pages 391–400. Springer Berlin Heidelberg.spa
dc.relation.referencesSiddiqi, K., Shokoufandeh, A., Dickinson, S. J., and Zucker, S. W. (1999). Shock graphs and shape matching. International Journal of Computer Vision, 35(1):13–32.spa
dc.relation.referencesSmeets, D., Fabry, T., Hermans, J., Vandermeulen, D., and Suetens, P. (2009). Isometric deformation modelling for object recognition. In Computer Analysis of Images and Patterns, pages 757–765. Springer Berlin Heidelberg.spa
dc.relation.referencesSobiecki, A., Jalba, A., and Telea, A. (2014). Comparison of curve and surface skeletonization methods for voxel shapes. Pattern Recognition Letters, 47:147–156.spa
dc.relation.referencesSobiecki, A., Yasan, H. C., Jalba, A. C., and Telea, A. C. (2013). Qualitative Comparison of Contraction- Based Curve Skeletonization Methods. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), volume 7883 LNCS, pages 425–439. Springer.spa
dc.relation.referencesStiene, S., Lingemann, K., Nuchter, A., and Hertzberg, J. (2006). Contour-based object detection in range images. In Third International Symposium on 3D Data Processing, Visualization, and Transmission (3DPVT’06), pages 168–175.spa
dc.relation.referencesStoyan, D. (1989). [a survey of the statistical theory of shape]: Comment. Statist. Sci., 4(2):115–116.spa
dc.relation.referencesSu, H., Maji, S., Kalogerakis, E., and Learned-Miller, E. (2015). Multi-view convolutional neural networks for 3d shape recognition. In 2015 IEEE International Conference on Computer Vision (ICCV), pages 945–953.spa
dc.relation.referencesSun, J., Ovsjanikov, M., and Guibas, L. (2009). A Concise and Provably Informative Multi-Scale Signature Based on Heat Diffusion. Computer Graphics Forum, 28(5):1383–1392.spa
dc.relation.referencesSuper, B. (2004). Learning chance probability functions for shape retrieval or classification. In Conference on Computer Vision and Pattern Recognition Workshop. IEEE.spa
dc.relation.referencesSuper, B. J. (2006). RETRIEVAL FROM SHAPE DATABASES USING CHANCE PROBABILITY FUNCTIONS AND FIXED CORRESPONDENCE. International Journal of Pattern Recognition and Artificial Intelligence, 20(08):1117–1137.spa
dc.relation.referencesTagliasacchi, A., Delame, T., Spagnuolo, M., Amenta, N., and Telea, A. (2016). 3d skeletons: A state-of- the-art report. Computer Graphics Forum, 35(2):573–597.spa
dc.relation.referencesTal, A. (2014). 3D Shape Analysis for Archaeology, pages 50–63. Springer Berlin Heidelberg, Berlin, Heidel- berg.spa
dc.relation.referencesThompson, D. W. (1942). On growth and form / by D’Arcy Wentworth Thompson. Cambridge University Press Cambridge, Eng, 2nd ed. edition.spa
dc.relation.referencesToshev, A. (2011). Shape Representations For Object Recognition. PhD thesis, University of Pennsylvania.spa
dc.relation.referencesToshev, A., Taskar, B., and Daniilidis, K. (2012). Shape-based object detection via boundary structure segmentation. International Journal of Computer Vision, 99(2):123–146.spa
dc.relation.referencesTsogkas, S. and Dickinson, S. J. (2017). Amat: Medial axis transform for natural images. 2017 IEEE International Conference on Computer Vision (ICCV), pages 2727–2736.spa
dc.relation.referencesTu, Z. and Yuille, A. L. (2004). Shape matching and recognition – using generative models and informative features. In Lecture Notes in Computer Science, pages 195–209. Springer Berlin Heidelberg.spa
dc.relation.referencesvan der Maaten, L. J. P., Boon, P. J., Lange, G., Paijmans, H., and Postma, E. O. (2006). Computer vision and machine learning for archaeology. In Proceedings of the Computer Applications in Archaeology, CAA 2006, page in press. Dr. H. Kamermans, Faculty of Archeology, Leiden University.spa
dc.relation.referencesViswanathan, G. K., Murugesan, A., and Nallaperumal, K. (2013). A parallel thinning algorithm for contour extraction and medial axis transform. In 2013 IEEE International Conference ON Emerging Trends in Computing, Communication and Nanotechnology (ICECCN), pages 606–610.spa
dc.relation.referencesWachinger, C., Salat, D. H., Weiner, M., and and, M. R. (2016). Whole-brain analysis reveals increased neuroanatomical asymmetries in dementia for hippocampus and amygdala. Brain, 139(12):3253–3266.spa
dc.relation.referencesWafi, N. M., Yaakob, S. N., Salim, N. S., Jusoh, M., Nazren, A. R. A., and Hisham, M. B. (2016). Im- age analysis using new descriptors average feature optimization based on fourier descriptors technique. In 2016 International Conference on Radar, Antenna, Microwave, Electronics, and Telecommunications (ICRAMET), pages 135–138.spa
dc.relation.referencesWang, C., Cheng, M., Sohel, F., Bennamoun, M., and Li, J. (2019a). NormalNet: A voxel-based CNN for 3d object classification and retrieval. Neurocomputing, 323:139–147.spa
dc.relation.referencesWang, Y., Sun, Y., Liu, Z., Sarma, S. E., Bronstein, M. M., and Solomon, J. M. (2019b). Dynamic graph cnn for learning on point clouds. ACM Transactions on Graphics (TOG).spa
dc.relation.referencesWang, Y., Xu, Y., Tsogkas, S., Bai, X., Dickinson, S. J., and Siddiqi, K. (2018). Deepflux for skeletons in the wild. ArXiv, abs/1811.12608.spa
dc.relation.referencesWorrall, D. E. and Brostow, G. J. (2018). Cubenet: Equivariance to 3d rotation and translation. CoRR, abs/1804.04458.spa
dc.relation.referencesWorrall, D. E., Garbin, S. J., Turmukhambetov, D., and Brostow, G. J. (2017). Harmonic networks: Deep translation and rotation equivariance. In 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE.spa
dc.relation.referencesXie, J., Heng, P.-A., and Shah, M. (2008). Shape matching and modeling using skeletal context. Pattern Recognition, 41(5):1756 – 1767.spa
dc.relation.referencesYang, S. and Wang, Y. (2007). Rotation invariant shape contexts based on feature-space fourier transfor- mation. In Fourth International Conference on Image and Graphics (ICIG 2007), pages 575–579.spa
dc.relation.referencesYang, X., Koknar-Tezel, S., and Latecki, L. J. (2009). Locally constrained diffusion process on locally densified distance spaces with applications to shape retrieval. In 2009 IEEE Conference on Computer Vision and Pattern Recognition. IEEE.spa
dc.relation.referencesYe, J. and Yu, Y. (2015). A fast modal space transform for robust nonrigid shape retrieval. The Visual Computer, 32(5):553–568.spa
dc.relation.referencesYe Mei and Androutsos, D. (2008). Affine invariant shape descriptors: The ica-fourier descriptor and the pca-fourier descriptor. In 2008 19th International Conference on Pattern Recognition, pages 1–4.spa
dc.relation.referencesZeng, A., Song, S., Nießner, M., Fisher, M., and Xiao, J. (2016). 3dmatch: Learning the matching of local 3d geometry in range scans. CoRR, abs/1603.08182.spa
dc.relation.referencesZhang, T. Y. and Suen, C. Y. (1984). A fast parallel algorithm for thinning digital patterns. Commun. ACM, 27(3):236–239.spa
dc.relation.referencesZhao, Y. and Belkasim, S. (2012). Multiresolution fourier descriptors for multiresolution shape analysis. IEEE Signal Processing Letters, 19(10):692–695.spa
dc.relation.referencesZhihu Huang and Jinsong Leng (2010). Analysis of hu’s moment invariants on image scaling and rotation. In 2010 2nd International Conference on Computer Engineering and Technology, volume 7, pages V7– 476–V7–480.spa
dc.relation.referencesZhirong Wu, Song, S., Khosla, A., Fisher Yu, Linguang Zhang, Xiaoou Tang, and Xiao, J. (2015). 3d shapenets: A deep representation for volumetric shapes. In 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 1912–1920.spa
dc.rights.accessrightsinfo:eu-repo/semantics/openAccessspa
dc.rights.licenseAtribución-NoComercial 4.0 Internacionalspa
dc.rights.urihttp://creativecommons.org/licenses/by-nc/4.0/spa
dc.subject.ddc000 - Ciencias de la computación, información y obras generales::003 - Sistemasspa
dc.subject.ddc590 - Animalesspa
dc.subject.lembMorfología (Zoología)
dc.subject.lembEsqueleto animal
dc.subject.proposalMedial Axis Transformeng
dc.subject.proposalIsometryeng
dc.subject.proposalMorphological Skeletonizationeng
dc.subject.proposalShape Analysis and Descriptioneng
dc.subject.proposalShape featureeng
dc.subject.proposalInvariance and Equivarianceeng
dc.subject.proposalPointNeteng
dc.subject.proposalChordiogameng
dc.subject.proposalShape Classification and Retrievaleng
dc.subject.proposalTransformada del eje mediospa
dc.subject.proposalIsometríaspa
dc.subject.proposalEsqueletos topológicosspa
dc.subject.proposalAnálisis y descripción de formaspa
dc.subject.proposalInvarianza y equivarianzaspa
dc.subject.proposalClasificación y recuperación de formasspa
dc.titleDescripción y análisis de forma basado en la invarianza a isometrías de los esqueletos topológicosspa
dc.title.translatedShape analysis and description based on the isometric invariances of topological skeletonizationeng
dc.typeTrabajo de grado - Doctoradospa
dc.type.coarhttp://purl.org/coar/resource_type/c_db06spa
dc.type.coarversionhttp://purl.org/coar/version/c_ab4af688f83e57aaspa
dc.type.contentTextspa
dc.type.driverinfo:eu-repo/semantics/doctoralThesisspa
dc.type.redcolhttp://purl.org/redcol/resource_type/TDspa
dc.type.versioninfo:eu-repo/semantics/acceptedVersionspa
dcterms.audienceEspecializadaspa
oaire.accessrightshttp://purl.org/coar/access_right/c_abf2spa

Archivos

Bloque original

Mostrando 1 - 1 de 1
Cargando...
Miniatura
Nombre:
1020415879.2019.pdf
Tamaño:
10.94 MB
Formato:
Adobe Portable Document Format
Descripción:
Tesis de Doctorado en Ingeniería - Sistemas

Bloque de licencias

Mostrando 1 - 1 de 1
No hay miniatura disponible
Nombre:
license.txt
Tamaño:
3.87 KB
Formato:
Item-specific license agreed upon to submission
Descripción: