scholarly journals Segmentation of Online Freehand Sketching Based on Speed Feature

2021 ◽  
Vol 2021 ◽  
pp. 1-11
Author(s):  
Guanfeng Wang ◽  
Shouxia Wang ◽  
Jingjing Kang ◽  
Shuxia Wang
Keyword(s):  
User Study ◽  
Feature Points ◽  
Sketch Recognition ◽  
Average Speed ◽  
Freehand Sketch ◽  
Novel Method ◽  
Geometric Primitives ◽  
Threshold Technique

We present a novel method to extract speed feature points for segmenting hand-drawn strokes into geometric primitives. The method consists of three steps. Firstly, the input strokes are classified into uniform and nonuniform speed strokes, representing a stroke drawn at relatively constant or uneven speeds, respectively. Then, a sharpening filter is used to enhance the peak features of the uniform speed strokes. Finally, a three-threshold technique that uses the average speed of the pen and its upper and lower deviations is used to extract speed feature points of strokes. We integrate the proposed method into our freehand sketch recognition (FSR) system to improve its robustness to support multiprimitive strokes. Through a user study with 8 participants, we demonstrate that the proposed method achieves higher segmentation efficiency in finding speed feature points than the existing method based on a single speed threshold.

The Evaluation of Direct Volume Rendering-Based Uncertainty Visualization Techniques for 3D Scalar Data

2014 ◽  
Vol 14 (04) ◽  
pp. 1450017
Author(s):  
Ji Ma ◽  
David Murphy ◽  
Gregory Provan ◽  
Cian O'Mathuna ◽  
Michael Hayes
Keyword(s):  
Volume Rendering ◽  
Completion Time ◽  
User Study ◽  
Task Completion ◽  
Direct Volume Rendering ◽  
Uncertainty Visualization ◽  
Task Completion Time ◽  
Scalar Data ◽  
Novel Method ◽  
Visualization Techniques

Many techniques have been proposed to represent uncertainty in data visualization. However, little research has been reported on the evaluation of their effectiveness. Moreover, no studies have been conducted to evaluate direct volume rendering (DVR)-based uncertainty visualization techniques. In this paper, we present a novel method that evaluates the perceptual effectiveness of four existing and one proposed DVR-based uncertainty visualization techniques. Four types of searching tasks that include identifying the maximum uncertainty data, identifying the minimum uncertainty data, identifying the maximum scalar data and identifying the minimum scalar data have been involved in this study, and a total of twenty-eight participants have contributed to the final main user study. Our analysis suggested that the proposed linked views and interactive specification (LVIS) technique appears to be the most accurate among all techniques, although it takes the longest task completion time. For the four existing techniques, the overlays technique appears to be the most advantageous, and it takes similar task completion time as the others. We believe that these findings can provide useful guidance for future uncertainty visualization design.

scholarly journals Research on Recognition Effect of DSCN Network Structure in Hand-Drawn Sketch

2021 ◽  
Vol 2021 ◽  
pp. 1-12
Author(s):  
Qunjing Ji
Keyword(s):  
Neural Network ◽  
Recognition Accuracy ◽  
Rapid Development ◽  
Image Features ◽  
Sketch Recognition ◽  
Freehand Sketch ◽  
Classical Models ◽  
Manual Selection ◽  
Stroke Sequence ◽  
Deep Learning Model

With the rapid development of image recognition technology, freehand sketch recognition has attracted more and more attention. How to achieve good recognition effect in the absence of color and texture information is the key to the development of freehand sketch recognition. Traditional nonlearning classical models are highly dependent on manual selection features. To solve this problem, a neural network sketch recognition method based on DSCN structure is proposed in this paper. Firstly, the stroke sequence of the sketch is drawn; then, the feature is extracted according to the stroke sequence combined with neural network, and the extracted image features are used as the input of the model to construct the time relationship between different image features. Through the control experiment on TU-Berlin dataset, the results show that, compared with the traditional nonlearning methods, HOG-SVM, SIFT-Fisher Vector, MKL-SVM, and FV-SP, the recognition accuracy of DSCN network is improved by 15.8%, 10.3%, 6.0%, and 2.9%, respectively. Compared with the classical deep learning model, Alex-Net, the recognition accuracy is improved by 5.6%. The above results show that the DSCN network proposed in this paper has strong ability of feature extraction and nonlinear expression and can effectively improve the recognition accuracy of hand-painted sketches after introducing the stroke order.

A Novel Method of Extracting Facial Feature Points Based on 3D Rotation and ASM

2011 ◽  
Vol 58-60 ◽  
pp. 1466-1470
Author(s):  
Jie Wu ◽  
Qing He ◽  
Ran Zhou ◽  
Chao Hu ◽  
Q. H. Meng
Keyword(s):  
Face Processing ◽  
Facial Feature ◽  
Face Image ◽  
Human Machine Interface ◽  
Active Shape Model ◽  
Final Position ◽  
Feature Points ◽  
Shape Model ◽  
Novel Method ◽  
Machine Interface

Precise extraction of facial feature points is fundamental to a variety of applications including face processing and human-machine interface. In this paper, a novel method of extracting facial feature points for profile faces is presented. This program is mainly based on a 3D rotation model of head and Active Shape Model (ASM). First we transform a profile face to a corresponding frontal face. Then, we implement the ASM program on the frontal face image. According to the relation between the profile face and frontal face, the final position of feature points on the profile face is obtained. We take limited facial feature points to do experiments and results show this kind of method is pretty effective.

A survey on freehand sketch recognition and retrieval

2019 ◽  
Vol 89 ◽  
pp. 67-87 ◽  
Author(s):  
Xianlin Zhang ◽  
Xueming Li ◽  
Yang Liu ◽  
Fangxiang Feng
Keyword(s):  
Sketch Recognition ◽  
Freehand Sketch

Smooth Edge Feature Lines Extraction from Point Clouds of Eroded Fractured Fragments

2013 ◽  
Vol 756-759 ◽  
pp. 4026-4030 ◽  
Author(s):  
Jian Bin Lin ◽  
Ming Quan Zhou ◽  
Zhong Ke Wu
Keyword(s):  
Principal Component Analysis ◽  
Principal Component ◽  
Point Clouds ◽  
Local Feature ◽  
Feature Points ◽  
Global Feature ◽  
Smooth Edge ◽  
Feature Lines ◽  
Novel Method ◽  
Edge Feature

This paper presents a novel method to extract edge lines from point clouds of these eroded, rough fractured fragments. Firstly, a principal component analysis based method is used to extract feature points, followed by clustering of these feature points. Secondly, a local feature lines fragment is constructed for each cluster and afterwards a smooth and noise pruning process for each local feature lines fragment. Thirdly, these separated local feature lines fragments are connected and bridged in order to eliminate the gaps caused by the eroded regions and construct completed global feature lines. Fourthly, a noise pruning process is performed. The output of this method is completed, smoothed edge feature lines. We illustrate the performance of our method on a number of real-world examples.

scholarly journals Sketch-Specific Data Augmentation for Freehand Sketch Recognition

2021 ◽  
Author(s):  
Ying Zheng ◽  
Hongxun Yao ◽  
Xiaoshuai Sun ◽  
Shengping Zhang ◽  
Sicheng Zhao ◽  
...  
Keyword(s):  
Data Augmentation ◽  
Sketch Recognition ◽  
Specific Data ◽  
Freehand Sketch

A Flexible Method for Reordering Feature Points on a Planar Target

2014 ◽  
Vol 543-547 ◽  
pp. 1179-1183
Author(s):  
Feng Jiao Li ◽  
Xiao Jing Li ◽  
Zhen Liu
Keyword(s):  
Machine Vision ◽  
Image Plane ◽  
Cross Ratio ◽  
Experimental Results ◽  
Feature Points ◽  
Foreign Objects ◽  
Novel Method ◽  
Planar Target

Planar targets have been widely used in the field of machine vision, and reordering the feature points on a planar target is of great difficulties and importance. As the current methods for that are of poor robustness, and are easily interfered by foreign objects or the image background, a novel method, which is of high versatility and is not easily affected by the interferences, is proposed in the paper. Cross ratio invariance and homographic relationship between the target plane and the image plane are utilized in the method. Experimental results show that the method is viable and robust to realize precise reordering of the feature points on planar targets in complex site environment.

Research of Automatic Positioning Technology for Dental Crowns Design

2010 ◽  
Vol 37-38 ◽  
pp. 953-956
Author(s):  
Yi Hao Duan ◽  
Xiao Sheng Cheng ◽  
Ning Dai ◽  
Tian Ran Yuan
Keyword(s):  
Three Dimensional ◽  
The Other ◽  
Feature Points ◽  
Dental Crowns ◽  
Novel Method ◽  
Automatic Positioning ◽  
Dental Crown

A novel method to automatically position the standard dental crown for the design of dental crowns is proposed. We use margin feature points to fit a plane, and the path of insertion is just a line perpendicular to the fitting plane, as well as through the centroid of the margin feature points. Besides, we can get the information of the matching points, one on the standard crown model and the other on the neighboring tooth model. According to these, the positioning datum is constructed. Finally through a series of three-dimensional transformation, automatic positioning of the standard dental crown can be implemented rapidly.

Surface Patch Primitive Based Object Modeling from CAD Data

2012 ◽  
Vol 162 ◽  
pp. 179-183 ◽  
Author(s):  
Sukhan Lee ◽  
Kyeong Dae Yoo ◽  
Jae Woong Kim ◽  
Moon Ju Lee
Keyword(s):  
Automated Reasoning ◽  
Direct Application ◽  
Robotic Manipulation ◽  
Surface Patch ◽  
Geometric Information ◽  
Grasp Planning ◽  
Manufacturing Automation ◽  
Novel Method ◽  
Robotic Automation ◽  
Geometric Primitives

For manufacturing automation, for instance, the robotic automation of automobile sub-assembly, CAD data serves as DB offering the geometric information of parts essential for robotic manipulation. However, a direct application of CAD for the robotic manipulation of parts may be of an issue, due to the fact that the particular format of the CAD data available, e.g., STL, does not directly provide certain geometric entities such as surface patch primitives and/or features that are required for robotic manipulation. In this paper, we present a novel method for extracting geometric primitives and/or features, such surface patch primitives as planar, cylindrical, conic, and spherical patches, from the STL format of CAD data, such that an industrial part/object can be represented as a logical sum of these surface patch primitives extracted. This surface patch primitive based modeling makes the automated reasoning involved in the recognition and pose estimation, as well as the grasp planning, of parts/objects easy to be done. The proposed method is applied to various CAD data samples for experimentation: the results demonstrate the reliability as well as the computational efficiency of the proposed method in the extraction of surface patch primitives.

Sign in / Sign up

Export Citation Format

Share Document