Filters








48,639 Hits in 3.2 sec

Performance Evaluation of Image Retrieval System using Non-parametric Techniques

Tranos Zuva, Seleman M. Ngwira, Sunday Olusegun Ojo, Keneilwe Zuva
2013 Journal of Multimedia Processing and Technologies  
Performance comparison of Epanechnikov, Gaussian and Histogram non-parametric algorithms was done in a generic image retrieval system.  ...  This paper considered performance of content based image retrieval system using image representation nonparametric algorithms.  ...  It is now difficult to search and retrieve the image of interest from this unstructured collection of images.  ... 
dblp:journals/jmpt/ZuvaNOZ13 fatcat:fiylt23ihfcg7i3vnklynuvvcm

3D Object Retrieval with Parametric Templates [article]

Roman Getto, Dieter W. Fellner
2015 Eurographics Workshop on 3D Object Retrieval, EG 3DOR  
We propose a 3D object retrieval system which uses parametric templates as prior knowledge for the retrieval.  ...  Our results show that the usage of parametric templates can raise the retrieval performance significantly.  ...  We have combined the parametric template with different descriptors and evaluated our approach with the Princeton Shape Benchmark, showing that the retrieval with a parametric template enhances the retrieval  ... 
doi:10.2312/3dor.20151054 fatcat:czp7ca6alveqhabbh6issk3wb4

Deep Parametric Shape Predictions using Distance Fields [article]

Dmitriy Smirnov, Matthew Fisher, Vladimir G. Kim, Richard Zhang, Justin Solomon
2020 arXiv   pre-print
Hence, we propose a new framework for predicting parametric shape primitives using deep learning.  ...  We use distance fields to transition between shape parameters like control points and input data on a pixel grid.  ...  Parametric shapes, on the other hand, use sparse sets of parameters like control points to express geometry.  ... 
arXiv:1904.08921v2 fatcat:4hgbzqp6rvhvdpcv572dzlk6wq

Deep Parametric Shape Predictions Using Distance Fields

Dmitriy Smirnov, Matthew Fisher, Vladimir G. Kim, Richard Zhang, Justin Solomon
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Hence, we propose a new framework for predicting parametric shape primitives using deep learning.  ...  We use distance fields to transition between shape parameters like control points and input data on a pixel grid.  ...  Parametric shapes, on the other hand, use sparse sets of parameters like control points to express geometry.  ... 
doi:10.1109/cvpr42600.2020.00064 dblp:conf/cvpr/0001FKZ020 fatcat:323tcj3eknc3rbcnqreamwdjti

Knowledge Base Question Answering by Case-based Reasoning over Subgraphs [article]

Rajarshi Das, Ameya Godbole, Ankita Naik, Elliot Tower, Robin Jia, Manzil Zaheer, Hannaneh Hajishirzi, Andrew McCallum
2022 arXiv   pre-print
We show that CBR-SUBG can answer queries requiring subgraph reasoning patterns and performs competitively with the best models on several KBQA benchmarks.  ...  other similar k-nearest neighbor (KNN) training queries along with query-specific subgraphs and (ii) a parametric component that is trained to identify the (latent) reasoning patterns from the subgraphs  ...  On comparison to parametric model.  ... 
arXiv:2202.10610v2 fatcat:xok2sbawizftnkulhufbgfar5m

Data-Driven Shape Analysis and Processing [article]

Kai Xu, Vladimir G. Kim, Qixing Huang, Evangelos Kalogerakis
2015 arXiv   pre-print
In addition, they are able to learn models that reason about properties and relationships of shapes without relying on hard-coded rules or explicitly programmed instructions.  ...  In contrast to traditional approaches, a key feature of data-driven approaches is that they aggregate information from a collection of shapes to improve the analysis and processing of individual shapes  ...  Given a data collection, the research focuses on grouping and relating data points, learning the data variations in the collection, and organizing the collection into a structured form, to facilitate retrieval  ... 
arXiv:1502.06686v1 fatcat:upajios4y5a6dgf2zw7faqai4a

An Adaptive Histogram Values & Texture Descriptor for Data Analysis

2016 International Journal of Science and Research (IJSR)  
Color, Texture, Shape and Layout are the primitive methods which are used in content based image retrieval system.  ...  In content based image retrieval system, images are stored by visual content, such as Texture, Shape, Layout and Color.  ...  The CBIR system retrieves relevant pics from an photograph collection primarily based on computerized derived functions.  ... 
doi:10.21275/v5i6.nov164646 fatcat:lrhicnn5vvgfhjmgii5wzd463i

Data-Driven Shape Analysis and Processing

Kai Xu, Vladimir G. Kim, Qixing Huang, Evangelos Kalogerakis
2016 Computer graphics forum (Print)  
Data-driven methods are also able to learn computational models that reason about properties and relationships of shapes without relying on hardcoded rules or explicitly programmed instructions.  ...  In contrast to traditional approaches that process shapes in isolation of each other, data-driven methods aggregate information from 3D model collections to improve the analysis, modelling and editing  ...  Existing approaches fall into two categories, based on how they represent the shape priors: parametric and non-parametric.  ... 
doi:10.1111/cgf.12790 fatcat:q76sq2syjvce5fjjjb7yoafww4

SketchOpt: Sketch-based Parametric Model Retrieval for Generative Design [article]

Mohammad Keshavarzi, Clayton Hutson, Chin-Yi Cheng, Mehdi Nourbakhsh, Michael Bergin, Mohammad Rahmani Asl
2020 arXiv   pre-print
Moreover, iterations of such models can be time-consuming tasks and sometimes limiting, as major changes in the layout design may result in remodeling the entire parametric definition.  ...  Developing fully parametric building models for performance-based generative design tasks often requires proficiency in many advanced 3D modeling and visual programming, limiting its use for many building  ...  [25] performed sketch-based shape retrieval by adopting a CNN architecture pre-trained on images, then fine-tuning it on a dataset of sketches collected by human volunteers [9] . Wang et al.  ... 
arXiv:2009.00261v1 fatcat:ujarwqsnvrew5ggc7f4mo3eeu4

UV-Net: Learning from Boundary Representations [article]

Pradeep Kumar Jayaraman, Aditya Sanghi, Joseph G. Lambourne, Karl D.D. Willis, Thomas Davies, Hooman Shayani, Nigel Morris
2021 arXiv   pre-print
Finally we demonstrate that UV-Net can generalize to supervised and unsupervised tasks on five datasets, while outperforming alternate 3D shape representations such as point clouds, voxels, and meshes.  ...  We introduce UV-Net, a novel neural network architecture and representation designed to operate directly on Boundary representation (B-rep) data from 3D CAD models.  ...  Deep Parametric Shape Processing.  ... 
arXiv:2006.10211v2 fatcat:vccqocfodvcihke2kk4lx6mvli

The Implications of 3D Thermal Structure on 1D Atmospheric Retrieval

Jasmina Blecic, Ian Dobbs-Dixon, Thomas Greene
2017 Astrophysical Journal  
When we use a higher resolution, more data points, and a parametrized temperature profile that allows more flexibility in the middle part of the atmosphere, we find a better match between the retrieved  ...  We also see some limitations in each of the temperature parametrizations.  ...  In Figure 19 we plotted possible profile shapes using this parametrization, allowing one parameter to vary and fixing the remaining parameters.  ... 
doi:10.3847/1538-4357/aa8171 fatcat:ghgkldtxfvdojlg35bgz3o5ftu

Data-driven shape analysis and processing

Kai Xu, Vladimir G. Kim, Qixing Huang, Niloy Mitra, Evangelos Kalogerakis
2016 SIGGRAPH ASIA 2016 Courses on - SA '16  
The success of data-driven approaches, backed by machine learning techniques, heavily relies on the accessibility of large data collections.  ...  As a result, knowledgedriven approaches are often hard to generalize well to large and diverse shape collections.  ...  Existing approaches fall into two categories, based on how they represent the shape priors: parametric and non-parametric.  ... 
doi:10.1145/2988458.2988473 dblp:conf/siggraph/0004KHMK16 fatcat:tefja76ijnclzmpux2iaj45zgu

Affine-invariant diffusion geometry for the analysis of deformable 3D shapes [article]

Dan Raviv, Alexander M. Bronstein, Michael M. Bronstein, Ron Kimmel, Nir Sochen
2010 arXiv   pre-print
Applications of the proposed framework demonstrate its power in generalizing and enriching the existing set of tools for shape analysis.  ...  The affine-invariant HKS descriptors can be used in one-to-one feature-based shape matching methods [23] , or in large-scale shape retrieval applications using the bags of features paradigm [17] .  ...  Performance of the proposed tools was demonstrated on shape retrieval, correspondence, and symmetry detection applications.  ... 
arXiv:1012.5933v1 fatcat:for64yoqu5ch3gk5dz3wgo666m

A 3-D Search engine based on Fourier series

Elmustapha Ait Lmaati, Ahmed El Oirrak, Driss Aboutajdine, Mohamed Daoudi, Mohammed Najib Kaddioui
2010 Computer Vision and Image Understanding  
A 3-D polygonal mesh model serves as a query for search by shape similarity in a large collection of 3-D models database using an interactive 3-D search engine.  ...  Therefore the development of recognition applications and retrieval systems of 3-D models is important. In this paper we propose a new scheme to measure similarity between 3-D models.  ...  Shape analysis should ideally be invariant to the choice of parameterizations, i.e. one should get the same comparison irrespective of parametrization of the two surfaces.  ... 
doi:10.1016/j.cviu.2009.09.010 fatcat:am7auzq6tjgsje7vnwtbx2rzqa

Integration of Explicit Geometric Constraints in the Comparison of 3D CAD Models for Part Design Reuse [chapter]

Antoine Brière-Côté, Louis Rivest, Roland Maranzana
2012 IFIP Advances in Information and Communication Technology  
While shape-based retrieval techniques contribute to the part design reuse process by retrieving similar part models, selecting the optimal candidate for reuse remains a challenge.  ...  We have developed a 3D CAD model comparison method focused on the representation of shape differences between similar models with respect to the reference's geometric constraints.  ...  One must therefore go back to the shape level representation to collect all the relevant details about the located differences.  ... 
doi:10.1007/978-3-642-35758-9_40 fatcat:ni7oumsem5evvn6jvl5nkohclm
« Previous Showing results 1 — 15 out of 48,639 results