arXiv.itsnick.co

an independent field guide to 23 million papers from arXiv & OpenAlex — not affiliated with arXiv.org

Author

Pin Xu

33 papers · 2009–2026

2009: 12010: 22011: 42012: 32013: 22014: 12015: 62016: 42017: 22018: 12019: 02020: 02021: 02022: 02023: 32024: 22025: 12026: 1
2009peak 6 in 20152026

Frequent co-authors

Papers

showing 1–25 of 33 RSS

Shot boundary detection in video retrieval

Zhonglan Wu, Pin Xu

Computer Vision and Pattern Recognition · · cited 17× · DOI ↗ · more like this ·

Video shot boundary detection is the basic part of content-based video retrieval, this paper elaborates the fundamental theory of the video shot boundary, and analyzes several typical shot boundary detection algorithms. Generally speaking, There are two kinds of shot change types: shot abrupt change and shot gradual change. Nowadays, the abrupt shot detection has been able to achieve a good result, but the result of the detection of gradual shot is still unsatisfying. This paper describes several typical algorithms in abrupt shot detection as well as in gradual shot detection. And then compare the results of several kinds of typical shot boundary detection algorithms.

A DRM Authentication Scheme Based on Smart-Card

Yichun Zhang, Lei Yang, Pin Xu +1 more

Information Systems · · cited 16× · DOI ↗ · more like this ·

In this paper, we proposed a DRM authentication scheme with smart-card. We first defined a three-role based DRM implementation scenario (server, playback device and smart-card), and then fully considered its security and performance challenges. Our scheme is based on identity based encryption and it's implementation-oriented. The scheme is efficient at server-side and compact in smart-card design. The proposed scheme realized mutual authentication and session key agreement among all involved parties and is immune to known potential attacks.

The relationship between work–family conflict and job satisfaction for preschool teachers in rural China: a moderated mediation model

Yingjie Wang, Qianqian Xia, Huilan Yue +5 more

Sociology and Political Science · Frontiers in Public Health · · cited 12× · DOI ↗ · more like this ·

Background: Job satisfaction for preschool teachers in rural areas has an important impact on their professional development, physical and mental health, and the development of preschool education. However, few studies have explored the factors that influence rural preschool teachers' job satisfaction. Purpose: This study aims to examine the influence of rural preschool teachers' work-family conflict on their job satisfaction, and the mediating effect of occupational identity, the moderating effect of social support. Method: Participants included 3,065 rural preschool teachers from Zhejiang Province in mainland China. Teachers completed questionnaires on work-family conflict, occupational identity, job satisfaction, and social support. The correlation and moderated mediation analyses were conducted using SPSS PROCESS. Results: (1) work-family conflict is associated with poorer job…

Research on the Technology of Video Key Frame Extraction Based on Clustering

Zhonglan Wu, Pin Xu

Computer Vision and Pattern Recognition · · cited 7× · DOI ↗ · more like this ·

The video key frame extraction technology is one of the important parts of content-based video retrieval. And the mainstream of key frame extraction is the algorithm based on clustering. The basic idea is: the video frames are grouped in accordance with the correlation of the visual content by clustering, and then we extract the most representative frame from each group as a key frame. In this paper, several common key frame extraction algorithms based on clustering are elaborated along with their advantages and disadvantages.

Visual tracking based on adaptive background modeling and improved particle filter

Xutang Li, Shanzhen Lan, Yue Jiang +1 more

Computer Vision and Pattern Recognition · · cited 5× · DOI ↗ · more like this ·

In this paper, a visual tracking algorithm based on adaptive background modeling and improved particle filter was proposed, which can be effectively used to track the players in sports video. Adaptive background modeling was used to extract motion information in the dynamic scene, and obtained precise motion region through a designed target selection filter, then, do the tracking algorithm under particle filter framework. The proposed algorithm solved the problem which needs to manually initialize the target position when tracking algorithm is applied to the engineering project. Besides, traditional resampling algorithm has degeneracy problem which could influence tracking effect when particle set is large, the paper also improved the resampling algorithm with a classified resample method. In the end, experimental results showed that the proposed algorithm has advantages in precision…

Real-Time 3D Facial Subtle Expression Control Based on Blended Normal Maps

Jiamin Gui, Yue Zhang, Shaobin Li +2 more

Computer Vision and Pattern Recognition · · cited 5× · DOI ↗ · more like this ·

Facial expression details can largely enhance the realistic facial animation. In this paper, we proposed a method to realize real-time control of 3D facial subtle expression, such as wrinkles and bulges, based on a set of normal maps. At runtime, we used a kinect to get the animation units for calculating control parameter. Normal blending technology was used to blend the base normal map and the detail normal map. A UV mask was combined with the blended normal map to achieve a good local control of facial subtle facial expression in real time.

Multiple feature fusion via hierarchical matching for TV logo recognition

Wenjie Chen, Shanzhen Lan, Pin Xu

Computer Vision and Pattern Recognition · · cited 5× · DOI ↗ · more like this ·

TV logo recognition plays an important role in video content understanding. So far, Real-time TV logo recognition under complex background based on single frame is still a very challenge task. By analyzing the characteristics of TV logo, a new TV logo recognition algorithm based on multiple feature fusion via hierarchical matching was proposed in this paper. The recognition process consisted of three stages. Firstly, a coarse matching based on frame differentiation and normalized cross correlation coefficient was employed to narrow down the candidate space significantly; then a fine matching based on HOG was used to describe the contours features of candidate logos; finally, the above features were fused to get better recognition performance. Experiment results showed the algorithm proposed had a remarkable performance in TV logo recognition under complex background based on single…

Objective Quality Assessment of Noised Stereoscopic Images

Yinghua Shen, Chaohui Lü, Pin Xu +1 more

Computer Vision and Pattern Recognition · · cited 5× · DOI ↗ · more like this ·

Stereoscopic images have been widely studied in recent years from a technical point of view, but the related quality assessment does not follow this enthusiasm. An objective quality assessment method for serious noised stereoscopic images is proposed. It is based on 2D quality evaluation and the characteristics of stereoscopic images. The disparity map for stereoscopic image pairs is an important indicator of stereoscopic image quality assessment. The metric Peak Signal to Noise Ratio (PSNR) is used to evaluate the quality of the noised disparity map, and the value of which is defined as DPSNR. The DPSNR methodology is used for objective quality assessment of serious noised stereoscopic images. The experimental results show that the proposed method could accurate and objective evaluation of stereoscopic images quality.

Stereoscopic Video Object Segmentation Based on Disparity Map

Yinghua Shen, Chaohui Lü, Pin Xu

Computer Vision and Pattern Recognition · · cited 5× · DOI ↗ · more like this ·

Stereoscopic video object segmentation plays an important role in stereo vision analysis and application. In this paper, a new method for disparity image segmentation derived from the extended mean shift theorem was proposed. Disparity estimation based on adaptive support weight approach was firstly introduced to get reliable depth information. Then an extended mean shift algorithm was exploited to achieve initial segmentation of depth image and coarse object areas were obtained. Finally, accurate objects were segmented according to the marking of other adjacent pixels. The experimental results show that the proposed method can segment the semantically meaningful objects successfully.

Improving DFU Image Classification by an Adaptive Augmentation Pool and Voting with Expertise

Xin Wu, Pin Xu, Haoyuan Chen +2 more

Computer Vision and Pattern Recognition · · cited 4× · DOI ↗ · more like this ·

Medical images are inseparable from data augmentation due to the lack of labeled data. However, several existing effective augments for natural images may not be suitable for medical images. This paper proposes an adaptive framework to construct a suitable augmentation pool for Diabetic Foot Ulcers medical images. Additionally, we use ensemble learning to enhance the model’s output. Instead of commonly used plurality voting, we propose a strategy named “voting with expertise” which prioritizes prediction with adequately reliable value. Experimental results show the effectiveness of the proposed methods and we won second place by combining the above two improvements in the ongoing open challenge-DFUC2021 Challenge.

A CNN based car model recognition improvement

Qi Bu, Shanzhen Lan, Pin Xu

Computer Vision and Pattern Recognition · · cited 4× · DOI ↗ · more like this ·

Car model recognition plays an important role in the road auto tolling system and the traffic volume statistics. In this paper, we present an improvement on car model recognition based on convolutional neural network. Firstly, our car image database which includes train and test images has been established. Then we preprocess our car model images using Gaussian Blur. Finally, we choose the appropriate CNN network structure Caffenet and improve the structure of it by adding an additional layer. The results indicate that through our series of improvement strategy, recognition accuracy has been improved to a better degree in terms of our database.

Real-Time Artistic Silhouettes Rendering for 3D Models

Dongliang Chen, Yue Zhang, Huimin Liu +1 more

Computer Graphics and Computer-Aided Design · · cited 4× · DOI ↗ · more like this ·

In line drawings, the line thickness is an important element for artists to depict shape and express emotion. In this paper, we proposed an approach for the real-time rendering of artistic silhouette outlines, based on the inherent aesthetic law of line thickness. We adopted the triangle shell technique as the basis of our silhouette outlines rendering method. By shifting the vertex of the back-facing mesh along its vertex normal under the aesthetic law, we recovered the variation of line thickness that artists often use when creating line drawings, and obtained aesthetically pleasing rendering results. Our method is implemented on GPU, and is suitable for real-time applications such as 3D games.

The detection of multi-objective billiards in snooker game video

Yan Ling, Shaobin Li, Pin Xu +1 more

Computer Vision and Pattern Recognition · · cited 4× · DOI ↗ · more like this ·

In this paper, we proposed a kind of multi-objective detection technique based on watershed in snooker game video. It can automatically extracts target markers without any man-machine interactions. As to target that has similar color with background, we make full use of its reflective region to detect it indirectly which reduces burden of detecting directly. Meanwhile, it can effectively separate gathered targets especially like gathered crimson balls. At first, for the single color of billiard table, we use color segmentation, Hough transform algorithm to extract pool table so as to get rid of the interference of other objects outside the pool table; Second, combine color segmentation with watershed algorithm to extract all non-green objectives; Then according to the relationship of reflective point and the target itself, we can pick up the green one. At last, we use the difference…

A judgment framework for roller skating using faster R-CNN

Qi Bu, Shanzhen Lan, Chaonan Gu +1 more

Computer Vision and Pattern Recognition · · cited 2× · DOI ↗ · more like this ·

Intelligent sports game referee is a new research direction in image processing and deep learning field. In this paper, we present an image processing framework for judgment of the speed slalom race in roller skating competition. Firstly, we chose faster R-CNN algorithm to identify targets such as pile and person during the race. Then an image processing framework is designed to process the recognized target in order to get the competition result. In this process, we remove the duplicate boxes after object recognition, modify the scale in RPN according to the target size and eliminate illumination influence by edge detection. After this, we complete our framework and experimental results show that the proposed framework is efficient and effective in terms of our database.

Illumination-Reflectance Based Image Enhancement Method for Character Recognition

Dongliang Chen, Shanzhen Lan, Pin Xu +1 more

Computer Vision and Pattern Recognition · · cited 2× · DOI ↗ · more like this ·

In order to improve the robustness of the existing character recognition systems and increase their recognition ratio under various bad illumination conditions, we proposed a fast image enhancement algorithm based on illumination-reflectance model, by using an approximate edge-preserved filter. Experimental results showed that our algorithm can greatly help increase the character recognition ratio. In addition, our algorithm can be expected to achieve real-time processing.

Real-time 3D facial expression control based on performance

Chen Chen, Yue Zhang, Pin Xu +3 more

Computer Vision and Pattern Recognition · · cited 2× · DOI ↗ · more like this ·

Performance-driven facial expression animation is one of the researching highlight in computer animation. In this paper, a method is proposed to control 3D facial expression animations in real time. The skeleton based on skinning was adopted for an efficient control of the deformation of the face model. Kinect was used as a capture device to get the depth and color images of the performers that help to obtain the animation units in candide-3. The animation unites smoothed by moving average filter were used to control the translation and rotation of skeletons, which produced a good result of the facial animations.

Stylized line rendering for three-dimensional models

Yue Zhang, Huimin Liu, Dongliang Chen +1 more

Computer Graphics and Computer-Aided Design · · cited 1× · DOI ↗ · more like this ·

Line rendering is an important issue in the field of Non-Photorealistic Rendering. How to use lines to describe the features of 3D models properly has long been a challenging goal for researchers, especially when some artistic rules have to be obeyed. In this article, a new method has been proposed for illustrating loose line style. We proposed a mathematical model according to the features of loose lines, our main work focus on three aspects as follows: silhouette artistic thickness performance, silhouette loose style display, and the temporal coherence of line texture performance.

3D Personalized Face Modeling Based on KINECT2

Jia Li, Yue Zhang, Pin Xu +2 more

Computer Vision and Pattern Recognition · · cited 1× · DOI ↗ · more like this ·

As a challenging topic in computer graphics and computer vision, 3D face modeling has been applied in various fields including film and animation production, game development as well as medical analysis. In this paper, we mainly aims at face personalized modeling optimization based on KINECT2. The optimization of the scheme proposed is the mesh deformation based on the differential coordinates constrained by face++ feature points and facial muscle function, namely firstly by extracting the location of feature points on the color image and depth map to obtain the feature information of the personalized face, then aligning the intermediate results of the KINECT2 modeling to the feature information. It is good to keep the local grid differential property, at the same time. In this paper, the automatic UV mapping is realized by means of planar differential mesh deformation, which improves…

Line Rendering in Loose Style for 3D Models

Dongliang Chen, Yue Zhang, Pin Xu

Computer Graphics and Computer-Aided Design · · cited 1× · DOI ↗ · more like this ·

The rendering of lines for 3D models is an important research topic in NPR(Non-Photorealistic Rendering). In line drawing, the use of loose style can help artists express their emotion and depict objects vividly. In this paper, we summarized the aesthetic factors of loose line style in two aspects: line thickness and line omission, and proposed a mathematical description for it. We also provided a real-time, temporal coherent rendering method for loose style lines. The subjective tests showed that our rendering results are sufficiently aesthetic pleasing.

Adaptive 3D facial deformation based on personal characteristics

Jia Li, Yue Zhang, Pin Xu +2 more

Computer Vision and Pattern Recognition · · cited 1× · DOI ↗ · more like this ·

With the vigorous development of 3D graphics and the Internet platform, the research of personal face deformation and face modeling of high quality is more and more important. In this paper, some experiments are made aiming at the process that is to create personalized model from the neutral face model by data matching and automatic deformation. We use Laplacian mesh editing as the mathematical method, and the deformation of facial mesh based on Laplacian differential coordinates is realized based on the personal characteristics obtained manually or from the photo.

Concatenated polar-coded multilevel modulation

Shengmei Zhao, Shun-pin Xu, Chao Xing

Computer Networks and Communications · · cited 1× · DOI ↗ · more like this ·

Combination of channel coding and modulation has been demonstrated to have high power and bandwidth efficiency. Compared with polar codes, the concatenated polar codes boost the error rate probability to almost exponential in the block-length with essentially no loss in computational complexity. In this paper, a concatenated polar-coded multilevel modulation scheme is proposed, where polar codes are the inner codes and Reed-Solomon or low density parity check codes are chosen as the outer codes. The bit error rate (BER) performances of the proposed scheme are analyzed for different inner polar code length and different inner polar code rate. The numerical simulation results show that the scheme has significantly improved in the BER performance in comparison with those results using multilevel polar-coded modulation and single level concatenated-polar coded modulation scheme. With the…

An Interactive Video Foreground Segmentation System Based on Modeling and Dynamic Graph Cut Algorithm

Xiao Yu Wu, Lei Yang, Shao Bin Li +1 more

Computer Vision and Pattern Recognition · Advanced materials research · · cited 1× · DOI ↗ · more like this ·

This paper proposes an interactive video foreground segmentation method based on modeling and graph cut algorithm. User interactions are required at initial frame or key frame of video sequence at first. Secondly we make use of user interactions information to develop background/foreground model and get foreground segmentation result of the current frame in term of graph cut algorithm. And automatic updated methods are proposed to obtain foreground segmentation results automatically on the later sequence of video without user interaction. The developed system of interactive video foreground segmentation has performances with extracting object and editing segmentation results. Experimental results on kinds of video demonstrated that our interactive segmentation system is efficient.

Effect of child-directed parent education for fresh parents in infant's psychological development.

Pin Xu

Sociology and Political Science · Chinese Journal of Child Health Care · · cited 1× · OpenAlex ↗ · more like this ·

【Objective】 To explore effects of child-directed parent education for fresh parents in children's psychology development. 【Methods】 420 children born from Sep.2007 to Jan.2008 in our hospital were enrolled into this study.These children were randomly selected and divided into two groups.Parents in observation group were given child-directed parent education,while others were not.Through questionnaire and interviews,the upbringing situation of all children were investgated.CDCC was used to assess the MDI,PDI of them.By using the Achenbach Child Behavior Checklist Standardized,all children's behaviors in this study were surveyed. 【Results】 1) The upbringing situation in observation group were significantly different with children in contral group(P0.05).2)There were no significant differences in mental development between two groups when these children were 6 months old(P0.05).While to 2…

Leveraging Sample Complementarity: A Novel Ensemble Strategy

Pin Xu, Yanyi Li, Yufei Han +3 more

Artificial Intelligence · · cited 0× · DOI ↗ · more like this ·

We propose a novel ensemble learning method based on sample complementarity that considers sample complementarity through scientific model selection and weight allocation. Our approach aims to capture the essence of ensemble learning, which is to achieve complementary advantages between models. Our ensemble method shares a similar level of simplicity in operation as averaging ensemble learning methods, while demonstrating better performance. We first select the model with the best performance to participate in the ensemble. Then, we identify and filter samples that show poor performance in the best model according to the performance to form a subset. We select models with strong complementarity for ensemble learning by evaluating the performance metrics of other models on this subset to determine their level of complementarity with the best model. Furthermore, we reasonably assign…

NMFA: Node-Level Multi-Frequency Adaptive Graph Neural Network for Graph Fraud Detection

Pin Xu, Youquan Wang, Mingjie Wu +1 more

Artificial Intelligence · · cited 0× · DOI ↗ · more like this ·

Fraudsters excel at concealing themselves by establishing connections with many trustworthy users to evade detectors. This cunning behavior increases the heterogeneity within the graph, rendering ineffective Graph Neural Network (GNN) methods based on homogeneity assumptions. To tackle this problem, several spectral GNN-based approaches have been introduced. However, existing methods fail to fully exploit frequency information, often overlooking signals of other frequencies in the graph, which diminishes the model's ability to capture node characteristics. Furthermore, many methods employ global filtering GNNs, neglecting the different frequencies of attention for fraudulent and benign nodes. To mitigate these shortcomings, we propose a new method called Node-Level Multi-Frequency Adaptive Graph Neural Network (NMFA). Specifically, our method is equipped with three multi-frequency…