scholarly journals Deep Common Semantic Space Embedding for Sketch-Based 3D Model Retrieval

Entropy ◽  
2019 ◽  
Vol 21 (4) ◽  
pp. 369 ◽  
Author(s):  
Jing Bai ◽  
Mengjie Wang ◽  
Dexin Kong
Keyword(s):  
Visual Perception ◽  
Computer Aided Design ◽  
3D Model ◽  
Semantic Content ◽  
3D Models ◽  
Semantic Space ◽  
3D Model Retrieval ◽  
Novel Approach ◽  
Model Retrieval ◽  
Aided Design

Sketch-based 3D model retrieval has become an important research topic in many applications, such as computer graphics and computer-aided design. Although sketches and 3D models have huge interdomain visual perception discrepancies, and sketches of the same object have remarkable intradomain visual perception diversity, the 3D models and sketches of the same class share common semantic content. Motivated by these findings, we propose a novel approach for sketch-based 3D model retrieval by constructing a deep common semantic space embedding using triplet network. First, a common data space is constructed by representing every 3D model as a group of views. Second, a common modality space is generated by translating views to sketches according to cross entropy evaluation. Third, a common semantic space embedding for two domains is learned based on a triplet network. Finally, based on the learned features of sketches and 3D models, four kinds of distance metrics between sketches and 3D models are designed, and sketch-based 3D model retrieval results are achieved. The experimental results using the Shape Retrieval Contest (SHREC) 2013 and SHREC 2014 datasets reveal the superiority of our proposed method over state-of-the-art methods.

scholarly journals An Efficient 3D Model Retrieval Method Based on Convolutional Neural Network

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-14
Author(s):  
Bo Ding ◽  
Lei Tang ◽  
Yong-jun He
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
3D Model ◽  
3D Models ◽  
3D Model Retrieval ◽  
Similarity Matching ◽  
Retrieval Method ◽  
Model Classification ◽  
Model Retrieval ◽  
Index Building

Recently, 3D model retrieval based on views has become a research hotspot. In this method, 3D models are represented as a collection of 2D projective views, which allows deep learning techniques to be used for 3D model classification and retrieval. However, current methods need improvements in both accuracy and efficiency. To solve these problems, we propose a new 3D model retrieval method, which includes index building and model retrieval. In the index building stage, 3D models in library are projected to generate a large number of views, and then representative views are selected and input into a well-learned convolutional neural network (CNN) to extract features. Next, the features are organized according to their labels to build indexes. In this stage, the views used for representing 3D models are reduced substantially on the premise of keeping enough information of 3D models. This method reduces the number of similarity matching by 87.8%. In retrieval, the 2D views of the input model are classified into a category with the CNN and voting algorithm, and then only the features of one category rather than all categories are chosen to perform similarity matching. In this way, the searching space for retrieval is reduced. In addition, the number of used views for retrieval is gradually increased. Once there is enough evidence to determine a 3D model, the retrieval process will be terminated ahead of time. The variable view matching method further reduces the number of similarity matching by 21.4%. Experiments on the rigid 3D model datasets ModelNet10 and ModelNet40 and the nonrigid 3D model dataset McGill10 show that the proposed method has achieved retrieval accuracy rates of 94%, 92%, and 100%, respectively.

scholarly journals A New 3D Model Retrieval Method with Building Blocks

2009 ◽  
Vol 2009 ◽  
pp. 1-6 ◽  
Author(s):  
Mingquan Zhou ◽  
Qingsong Huo ◽  
Guohua Geng ◽  
Xiaojing Liu
Keyword(s):  
3D Model ◽  
Building Blocks ◽  
3D Models ◽  
3D Model Retrieval ◽  
Virtual Space ◽  
3D Data ◽  
Model Retrieval ◽  
Voxel Model ◽  
Interactive 3D ◽  
Application Fields

As the numbers of 3D models available grow in many application fields, there is an increasing need for a search method to help people find them. Unfortunately, traditional search techniques are not always effective for 3D data. In this paper, we describe a novel method of interactive 3D model retrieval with building blocks. First, by using a cube block as the baseblock in a 3D virtual space, we may construct the query model with human-computer interaction method. Then through retrieving the polygon model of the database generated by the voxel model, we may get retrieval results in real time. Experiments are conducted to evaluate the performance of the proposed method.

scholarly journals Problems of Creation and Usage of 3D Model of Structures and Theirs Possible Solution

Symmetry ◽  
2020 ◽  
Vol 12 (1) ◽  
pp. 181 ◽  
Author(s):  
Dalibor Bartonek ◽  
Michal Buday
Keyword(s):  
Computer Aided Design ◽  
3D Model ◽  
Three Dimensional ◽  
3D Models ◽  
Cad System ◽  
Building Models ◽  
Special Cad ◽  
Aided Design ◽  
The Church

This article describes problems that occur when creating three-dimensional (3D) building models. The first problem is geometric accuracy; the next is the quality of visualization of the resulting model. The main cause of this situation is that current Computer-Aided Design (CAD) software does not have sufficient means to precision mapping the measured data of a given object in field. Therefore the process of 3D model creation is mainly a relatively high proportion of manual work when connecting individual points, approximating curves and surfaces, or laying textures on surfaces. In some cases, it is necessary to generalize the model in the CAD system, which degrades the accuracy and quality of field data. The article analyzes these problems and then recommends several variants for their solution. There are described two basic methods: using topological codes in the list of coordinates points and creating new special CAD features while using Python scripts. These problems are demonstrated on examples of 3D models in practice. These are mainly historical buildings in different locations and different designs (brick or wooden structures). These are four sacral buildings in the Czech Republic (CR): the church of saints Johns of Brno-Bystrc, the Church of St. Paraskiva in Blansko, further the Strejc’s Church in Židlochovice, and Church of St. Peter in Alcantara in Karviná city. All of the buildings were geodetically surveyed by terrestrial method while using total station. The 3D model was created in both cases in the program AUTOCAD v. 18 and MicroStation.

scholarly journals Cube of Space Sampling for 3D Model Retrieval

2021 ◽  
Vol 11 (23) ◽  
pp. 11142
Author(s):  
Zong-Yao Chen ◽  
Chih-Fong Tsai ◽  
Wei-Chao Lin
Keyword(s):  
3D Model ◽  
State Of The Art ◽  
3D Models ◽  
3D Model Retrieval ◽  
Feature Representation ◽  
Direct Extraction ◽  
Feature Descriptors ◽  
Model Retrieval ◽  
Representation Method ◽  
Global And Local

Since the number of 3D models is rapidly increasing, extracting better feature descriptors to represent 3D models is very challenging for effective 3D model retrieval. There are some problems in existing 3D model representation approaches. For example, many of them focus on the direct extraction of features or transforming 3D models into 2D images for feature extraction, which cannot effectively represent 3D models. In this paper, we propose a novel 3D model feature representation method that is a kind of voxelization method. It is based on the space-based concept, namely CSS (Cube of Space Sampling). The CSS method uses cube space 3D model sampling to extract global and local features of 3D models. The experiments using the ESB dataset show that the proposed method to extract the voxel-based features can provide better classification accuracy than SVM and comparable retrieval results using the state-of-the-art 3D model feature representation method.

Cataloging the 3D web: the availability of educational 3D models on the internet

2016 ◽  
Vol 34 (2) ◽  
pp. 239-258 ◽  
Author(s):  
Michael Groenendyk
Keyword(s):  
Computer Aided Design ◽  
3D Model ◽  
3D Models ◽  
3D Modelling ◽  
The Internet ◽  
Content Type ◽  
3D Printers ◽  
3D Printed ◽  
The Subject ◽  
Aided Design

Purpose – The number of 3D models available on the internet to both students and educators is rapidly expanding. Not only are the 3D model collections of popular websites like Thingiverse.com growing, organizations such as the Smithsonian Institution and NASA have also recently begun to build collections of 3D models and make these openly accessible online. Yet, even with increased interest in 3D printing and 3D scanning technologies, little is known about the overall structure of the 3D models available on the internet. The paper aims to discuss this issue. Design/methodology/approach – To initiate this project, a list was built of 33 of the most widely used 3D model websites on the internet. Freely downloadable models, as well as models available for purchase or as 3D printed objects were included in the list. Once the list of 33 websites was created, the data for each individual 3D model in the collections was manually assembled and recorded. The titles of the 3D models, keywords, subject headings, license information, and number of views and downloads were recorded, as this information was available. The data were gathered between January and May 2015, and compiled into a CSV database. To determine how online 3D model content relates to a variety of educational disciplines, relevant subject terms for a variety of educational disciples were extracted from the EBSCO database system. With this list of subject terms in hand, the keywords in the CSV database of model information were searched for each of the subject terms, with an automated process using a Perl script. Findings – There have been many teachers, professors, librarians and students who have purchased 3D printers with little or no 3D modelling skills. Without these skills the owners of these 3D printers are entirely reliant on the content created and freely shared by others to make use of their 3D printers. As the data collected for this research paper shows, the vast majority of open 3D model content available online pertains to the professions already well versed in 3D modelling and Computer Aided Design design, such as engineering and architecture. Originality/value – Despite that fact that librarians, teachers and other educators are increasingly using technologies that rely on open 3D model content as educational tools, no research has yet been done to assess the number of 3D models available online and what educational disciplines this content relates to. This paper attempts to fill this gap, providing an overview of the size of this content, the educational disciplines this content relates to and who has so far been responsible for developing this content. This information will be valuable to librarians and teachers currently working with technology such as 3D printers and virtual reality, as well as those considering investing in this technology.

Sketch-Based 3D Model Retrieval Using Attributes

2018 ◽  
Vol 10 (3) ◽  
pp. 60-75 ◽  
Author(s):  
Haopeng Lei ◽  
Guoliang Luo ◽  
Yuhua Li ◽  
Jianming Liu ◽  
Jihua Ye
Keyword(s):  
3D Model ◽  
3D Models ◽  
3D Model Retrieval ◽  
Superior Performance ◽  
Visual Features ◽  
Low Level ◽  
Feature Descriptors ◽  
Model Retrieval ◽  
Sketch Retrieval ◽  
Semantic Attributes

With the rapid growth of available 3D models on the Internet, how to retrieve 3D models based on hand-drawn sketch retrieval are becoming increasingly important. This article proposes a new sketch-based 3D model retrieval approach. This approach is different from current methods that make use of low-level visual features to capture the search intention of users. The proposed method uses two kinds of semantic attributes, including pre-defined attributes and latent attributes. Specifically, pre-defined attributes are defined manually which can provide prior knowledge about different sketch categories and latent-attributes are more discriminative which can differentiate sketch categories at a finer level. Therefore, these semantic attributes can provide a more descriptive and discriminative meaningful representation than low-level feature descriptors. The experiment results demonstrate that this proposed method can achieve superior performance over previously proposed sketch-based 3D model retrieval methods.

Creative Exploration of Scaled Product Family 3D Models Using Gesture Based Conceptual Computer Aided Design (C-CAD) Tool

2013 ◽  
Author(s):  
Prakhar Jaiswal ◽  
Anurag Baburao Bajad ◽  
Vishwas Grama Nanjundaswamy ◽  
Anoop Verma ◽  
Rahul Rai
Keyword(s):  
Computer Aided Design ◽  
3D Model ◽  
Product Family ◽  
3D Models ◽  
Product Family Design ◽  
Modeling Tools ◽  
Design Exploration ◽  
Computer Aided ◽  
Aided Design ◽  
Family Models

The concepts of scale and platform based product family are being used by many companies to meet the customization needs of customers. Research in the area of product family design predominantly focuses on optimization frameworks. There is a lack of creative conceptual computer-aided 3D modeling tools for product family design exploration. In this paper, a gesture-based conceptual computer-aided design (C-CAD) exploration tool for scaled 3D product family models is presented. The proposed gesture-based C-CAD allows for easy, natural, and intuitive modification of 3D objects to create scaled 3D product family models. The input 3D model for conceptual design exploration purposes are obtained in two ways: (a) 3D model is generated by scanning an existing product using depth sensing (RGB-D) camera, and (b) 3D models available in large online repository such as Google Warehouse and TurboSquid are used as-is. Hand gestures recognized using the DepthSense® 311 (RGB-D) camera from SoftKinetic® are used in conjunction with Principal Component Analysis (PCA) based geometric algorithms to enable the interactive scaling of inputted 3D models. The efficacy of the proposed method is demonstrated through multiple example problems. The proposed method of 3D model exploration is most useful for product designs that are scaled variants.

A Novel 3D Model Retrieval Approach Based on Model-Partitioning and Fuzzy Relevance Feedback

2010 ◽  
Vol 143-144 ◽  
pp. 186-190
Author(s):  
Kuan Sheng Zou ◽  
Chun Ho Wu ◽  
Wai Hung Ip ◽  
Ching Yuen Chan ◽  
Kei Leung Yung ◽  
...  
Keyword(s):  
3D Model ◽  
Retrieval System ◽  
3D Models ◽  
3D Model Retrieval ◽  
Retrieval Performance ◽  
System A ◽  
Model Retrieval ◽  
Euclidean Motion ◽  
Feedback Method ◽  
Model Partitioning

3D models play an important role in many applications, so there is an urgent need for an effective content based 3D model retrieval system. A variety of 3D model retrieval methods have been proposed in recent years. Shape distributions show superiority over other methods due to ease of computation and invariance to Euclidean motion, but there is poor retrieval performance for the loss in information. This paper introduces two model-partitioning methods to improve shape distributions, in which the two enhanced descriptors are combined with a fuzzy feedback method. Experimental results show that the proposed methods can achieve better retrieval performance.

3D Model Retrieval Based on Distance Classification Histogram

2015 ◽  
Vol 733 ◽  
pp. 931-934 ◽  
Author(s):  
Ji Lai Zhou ◽  
Ming Quan Zhou ◽  
Guo Hua Geng
Keyword(s):  
3D Model ◽  
3D Models ◽  
3D Model Retrieval ◽  
Experimental Results ◽  
Model Surface ◽  
Model Retrieval

This paper presents a new algorithm to retrieve 3D model on distance classification histogram. First, we select the certain number of random points on the model surface and compute the distance between two random points. Secondly, we sort the distance into two types which is based on the different geometry properties of these distance and construct the distance classification histogram. Finally, we measure the similarity of 3D models by comparing distance classification histogram. The experimental results on PSB show that our method has a good performance in precision and computational complication.

Sign in / Sign up

Export Citation Format

Share Document