Dynamic character graph via online face clustering for movie analysis

Abstract An effective approach to automated movie content analysis involves building a network (graph) of its characters. Existing work usually builds a static character graph to summarize the content using metadata, scripts or manual annotations. We propose an unsupervised approach to building a dynamic character graph that captures the temporal evolution of character interaction. We refer to this as the character interaction graph (CIG). Our approach has two components: (i) an online face clustering algorithm that discovers the characters in the video stream as they appear, and (ii) simultaneous creation of a CIG using the temporal dynamics of the resulting clusters. We demonstrate the usefulness of the CIG for two movie analysis tasks: narrative structure (acts) segmentation and major character retrieval. Our evaluation on full-length movies containing more than 5000 face tracks shows that the proposed approach achieves superior performance for both the tasks.

Download Full-text

Multitarget Tracking Algorithm Based on Adaptive Network Graph Segmentation in the Presence of Measurement Origin Uncertainty

Sensors ◽

10.3390/s18113791 ◽

2018 ◽

Vol 18 (11) ◽

pp. 3791

Author(s):

Tianli Ma ◽

Song Gao ◽

Chaobo Chen ◽

Xiaoru Song

Keyword(s):

Shortest Path ◽

Clustering Algorithm ◽

Search Algorithm ◽

Optimal Solution ◽

Tracking Algorithm ◽

Multitarget Tracking ◽

Integer Programming Problem ◽

Tracking Accuracy ◽

Network Graph ◽

Adaptive Network

To deal with the problem of multitarget tracking with measurement origin uncertainty, the paper presents a multitarget tracking algorithm based on Adaptive Network Graph Segmentation (ANGS). The multitarget tracking is firstly formulated as an Integer Programming problem for finding the maximum a posterior probability in a cost flow network. Then, a network structure is partitioned using an Adaptive Spectral Clustering algorithm based on the Nyström Method. In order to obtain the global optimal solution, the parallel A* search algorithm is used to process each sub-network. Moreover, the trajectory set is extracted by the Track Mosaic technique and Rauch–Tung–Striebel (RTS) smoother. Finally, the simulation results achieved for different clutter intensity indicate that the proposed algorithm has better tracking accuracy and robustness compared with the A* search algorithm, the successive shortest-path (SSP) algorithm and the shortest path faster (SPFA) algorithm.

Download Full-text

Network analysis as a method of assessing terms in a dataset comprising eating disorders

International Journal of Engineering & Technology ◽

10.14419/ijet.v7i2.7.11079 ◽

2018 ◽

Vol 7 (2.7) ◽

pp. 841

Author(s):

Dr Adimulam Yesu Babu ◽

Dr Deepak Nedunuri ◽

T Venkata Sai Krishna

Keyword(s):

Eating Disorders ◽

Network Analysis ◽

Clustering Algorithm ◽

Agglomerative Hierarchical Clustering ◽

Network Graph ◽

Undirected Network ◽

Ward Clustering ◽

R Programming ◽

Network Centralities ◽

Word Frequencies

Eating disorders are central reason of physical and psycho-social morbidity. Several factors have been identified as being associated with the prevalence and progression of eating disorders in humans. Scientific investigation was carried out to assess the usage of terms in manuscript titles of nearly 900 published articles followed by network analysis and network centralities using R programming. The tm package, term document matrix function was utilized to create a term document matrix (TDM) from the corpus. A binary word matrix comprising 17 terms was created based on higher probability of occurring a term in a column. An agglomerative hierarchical clustering technique using ward clustering algorithm was presented. A data frame from the TDM was created to store data and used to plot word cloud based on word frequencies. An undirected network graph was plotted based on terms that appeared in the term matrix. Centralization measures such as Degree centrality, Closeness, Eigenvector and betweenness Centrality were reported.

Download Full-text

Simple primitive recognition via hierarchical face clustering

Computational Visual Media ◽

10.1007/s41095-020-0192-6 ◽

2020 ◽

Vol 6 (4) ◽

pp. 431-443

Author(s):

Xiaolong Yang ◽

Xiaohong Jia

Keyword(s):

Hierarchical Clustering ◽

Test Data ◽

Efficient Algorithm ◽

Clustering Algorithm ◽

State Of The Art ◽

Experimental Results ◽

Face Clustering ◽

Wide Range ◽

Hierarchical Clustering Algorithm ◽

Bottom To Top

AbstractWe present a simple yet efficient algorithm for recognizing simple quadric primitives (plane, sphere, cylinder, cone) from triangular meshes. Our approach is an improved version of a previous hierarchical clustering algorithm, which performs pairwise clustering of triangle patches from bottom to top. The key contributions of our approach include a strategy for priority and fidelity consideration of the detected primitives, and a scheme for boundary smoothness between adjacent clusters. Experimental results demonstrate that the proposed method produces qualitatively and quantitatively better results than representative state-of-the-art methods on a wide range of test data.

Download Full-text

Recognition of Online Handwritten Bangla Characters Using Supervised and Unsupervised Learning Approaches

International Journal of Computer Vision and Image Processing ◽

10.4018/ijcvip.2020070102 ◽

2020 ◽

Vol 10 (3) ◽

pp. 18-30

Author(s):

Prosenjit Mukherjee ◽

Shibaprasad Sen ◽

Kaushik Roy ◽

Ram Sarkar

Keyword(s):

Character Recognition ◽

Supervised Classification ◽

Clustering Algorithm ◽

Feature Vector ◽

Recognition Accuracy ◽

Learning Approaches ◽

Current Experiment ◽

Unsupervised Approach ◽

Mean Structure ◽

Euclidian Distances

This paper explores the domain of online handwritten Bangla character recognition by stroke-based approach. The component strokes of a character sample are recognized firstly and then characters are constructed from the recognized strokes. In the current experiment, strokes are recognized by both supervised and unsupervised approaches. To estimate the features, images of all the component strokes are superimposed. A mean structure has been generated from this superimposed image. Euclidian distances between pixel points of a stroke sample and mean stroke structure are considered as features. For unsupervised approach, K-means clustering algorithm has been used whereas six popular classifiers have been used for supervised approach. The proposed feature vector has been evaluated on 10,000-character database and achieved 90.69% and 97.22% stroke recognition accuracy in unsupervised (using K-means clustering) and supervised way (using MLP [multilayer perceptron] classifier). This paper also discusses about merit and demerits of unsupervised and supervised classification approaches.

Download Full-text

Supervised clustering of high-dimensional data using regularized mixture modeling

Briefings in Bioinformatics ◽

10.1093/bib/bbaa291 ◽

2020 ◽

Author(s):

Wennan Chang ◽

Changlin Wan ◽

Yong Zang ◽

Chi Zhang ◽

Sha Cao

Keyword(s):

Clustering Algorithm ◽

Expectation Maximization Algorithm ◽

Substantial Improvement ◽

Superior Performance ◽

High Dimensional ◽

Analysis Tool ◽

Supervised Clustering ◽

Mixture Regression ◽

Clustering Problem ◽

Clinical Presentations

Abstract Identifying relationships between genetic variations and their clinical presentations has been challenged by the heterogeneous causes of a disease. It is imperative to unveil the relationship between the high-dimensional genetic manifestations and the clinical presentations, while taking into account the possible heterogeneity of the study subjects.We proposed a novel supervised clustering algorithm using penalized mixture regression model, called component-wise sparse mixture regression (CSMR), to deal with the challenges in studying the heterogeneous relationships between high-dimensional genetic features and a phenotype. The algorithm was adapted from the classification expectation maximization algorithm, which offers a novel supervised solution to the clustering problem, with substantial improvement on both the computational efficiency and biological interpretability. Experimental evaluation on simulated benchmark datasets demonstrated that the CSMR can accurately identify the subspaces on which subset of features are explanatory to the response variables, and it outperformed the baseline methods. Application of CSMR on a drug sensitivity dataset again demonstrated the superior performance of CSMR over the others, where CSMR is powerful in recapitulating the distinct subgroups hidden in the pool of cell lines with regards to their coping mechanisms to different drugs. CSMR represents a big data analysis tool with the potential to resolve the complexity of translating the clinical representations of the disease to the real causes underpinning it. We believe that it will bring new understanding to the molecular basis of a disease and could be of special relevance in the growing field of personalized medicine.

Download Full-text

Adaptive Initialization Method for K-Means Algorithm

Frontiers in Artificial Intelligence ◽

10.3389/frai.2021.740817 ◽

2021 ◽

Vol 4 ◽

Author(s):

Jie Yang ◽

Yu-Kai Wang ◽

Xin Yao ◽

Chin-Teng Lin

Keyword(s):

Time Complexity ◽

Clustering Algorithm ◽

Clustering Algorithms ◽

Real Life ◽

Superior Performance ◽

Local Optima ◽

Initial Cluster ◽

Higher Dimensional ◽

Real World Datasets ◽

Random Method

The K-means algorithm is a widely used clustering algorithm that offers simplicity and efficiency. However, the traditional K-means algorithm uses a random method to determine the initial cluster centers, which make clustering results prone to local optima and then result in worse clustering performance. In this research, we propose an adaptive initialization method for the K-means algorithm (AIMK) which can adapt to the various characteristics in different datasets and obtain better clustering performance with stable results. For larger or higher-dimensional datasets, we even leverage random sampling in AIMK (name as AIMK-RS) to reduce the time complexity. 22 real-world datasets were applied for performance comparisons. The experimental results show AIMK and AIMK-RS outperform the current initialization methods and several well-known clustering algorithms. Specifically, AIMK-RS can significantly reduce the time complexity to O (n). Moreover, we exploit AIMK to initialize K-medoids and spectral clustering, and better performance is also explored. The above results demonstrate superior performance and good scalability by AIMK or AIMK-RS. In the future, we would like to apply AIMK to more partition-based clustering algorithms to solve real-life practical problems.

Download Full-text

A Bayesian Dynamical Approach for Human Action Recognition

Sensors ◽

10.3390/s21165613 ◽

2021 ◽

Vol 21 (16) ◽

pp. 5613

Author(s):

Amirreza Farnoosh ◽

Zhouping Wang ◽

Shaotong Zhu ◽

Sarah Ostadabbas

Keyword(s):

Action Recognition ◽

Large Scale ◽

Temporal Dynamics ◽

Human Action Recognition ◽

Human Action ◽

Superior Performance ◽

Action Classification ◽

Motion Data ◽

Highly Correlated ◽

Low Dimensional

We introduce a generative Bayesian switching dynamical model for action recognition in 3D skeletal data. Our model encodes highly correlated skeletal data into a few sets of low-dimensional switching temporal processes and from there decodes to the motion data and their associated action labels. We parameterize these temporal processes with regard to a switching deep autoregressive prior to accommodate both multimodal and higher-order nonlinear inter-dependencies. This results in a dynamical deep generative latent model that parses meaningful intrinsic states in skeletal dynamics and enables action recognition. These sequences of states provide visual and quantitative interpretations about motion primitives that gave rise to each action class, which have not been explored previously. In contrast to previous works, which often overlook temporal dynamics, our method explicitly model temporal transitions and is generative. Our experiments on two large-scale 3D skeletal datasets substantiate the superior performance of our model in comparison with the state-of-the-art methods. Specifically, our method achieved 6.3% higher action classification accuracy (by incorporating a dynamical generative framework), and 3.5% better predictive error (by employing a nonlinear second-order dynamical transition model) when compared with the best-performing competitors.

Download Full-text

Analysis of Human-Land Coupled Bearing Capacity of Qiangtang Meadow in Northern Tibet Based on Fuzzy Clustering Algorithm

Mathematical Problems in Engineering ◽

10.1155/2020/8231237 ◽

2020 ◽

Vol 2020 ◽

pp. 1-9

Author(s):

ShiBao Li

Keyword(s):

Energy Consumption ◽

Bearing Capacity ◽

Fuzzy Clustering ◽

Clustering Algorithm ◽

High Energy ◽

Geographic Information ◽

Superior Performance ◽

Capacity Analysis ◽

Northern Tibet ◽

Fuzzy Clustering Algorithm

Aiming to address the problems of high energy consumption, low efficiency, low correlation between the analyzed and actual results, and poor rationality of research indexes in current methods of analysis of human-land coupled bearing capacity of meadows, a novel method of human-land coupled bearing capacity analysis of Qiangtang meadow in northern Tibet, based on fuzzy clustering algorithm, is proposed. Basic geographic information data in Tibet were acquired, the collected data images were registered by ENVI4.2 software, and the collected data were vectorized by ArcGIS 9.3 software to construct a basic geographic information database in Tibet. Based on the frequency domain processing algorithm, the geographic information image was suppressed by noise and filtered by using a high-pass filter to realize the geographic information data processing in the study area. The human-land coupled bearing capacity analysis of Qiangtang meadow in northern Tibet was evaluated through fuzzy clustering, bearing capacity evaluation, and bearing capacity calculation under the sharing of closure. The experimental results showed that the average running energy consumption of the method was 81 J, and 97% of the analyzed results were consistent with the actual situation. These results indicate that the operation efficiency of the method is high, and the rationality coefficient of the research index is large. The proposed method has superior performance and feasibility.

Download Full-text

Wireless Sensor Network Coverage Hole Localization by Ant Colony Optimized Gaussian Mixture Model Clustering

Journal of Computational and Theoretical Nanoscience ◽

10.1166/jctn.2020.8920 ◽

2020 ◽

Vol 17 (6) ◽

pp. 2488-2495

Author(s):

Shalu ◽

Amita Malik

Keyword(s):

Wireless Sensor Network ◽

Sensor Network ◽

Gaussian Mixture Model ◽

Mixture Model ◽

Clustering Algorithm ◽

Gaussian Mixture ◽

Wireless Sensor ◽

Network Graph ◽

Graph Metrics ◽

Coverage Hole

Nodes in the wireless sensor network have a minimal power source and they exhaust very quickly in communicating with each other. If any of the nodes die, a coverage hole creates in that region. This coverage hole leads to fast energy depletion of other nodes along with the security issues due to intruder node’s placement at that location. The solution to detection of coverage hole is discussed in our paper and it is experimentally validated. We propose an unsupervised machine learning clustering algorithm to cluster the network graph metrics. An undirected network graph of nodes is created and five graph metrics are extracted. The vector of features is clustered by Ant colony optimized expectation.maximization Gaussian mixture model (ACO-EM GMM) clustering algorithm. Our algorithm is compared with the state of art works based on false detection parameter.

Download Full-text

A medical network clustering method with weighted graph structure

Measurement and Control ◽

10.1177/0020294020952469 ◽

2020 ◽

pp. 002029402095246

Author(s):

Hong Wu ◽

Zijian Fu ◽

Yizhou Wang

Keyword(s):

Clustering Algorithm ◽

Weighted Graph ◽

Real Data ◽

Diffusion Method ◽

Network Clustering ◽

Graph Structure ◽

Network Graph ◽

Analysis Algorithm ◽

Network Function ◽

Medical Network

Today, most of the databases used for drug information mining are derived from the collection of many treatments under a single disease, and some special drug compatibility rules can be found from them. However, researchers’ exploration of medical data is not limited to this. The comparative analysis of drugs for different diseases has become a new research point. In this paper, the drug is used as a node, the relationship is the edge connecting the two nodes, the co-occurrence frequency of the drug is used as the weight of the edge to establish a network graph. We use the clustering algorithm of the weighted network graph center diffusion method combining the network topology and the edge weights to divide the network graph into communities. Then we proposed the Structural Clustering Algorithm on Weighted Networks (SCW), it helps to study the prescription of medical prescriptions and provides more scientific recommendations for auxiliary prescriptions. In the experiment, SCW is compared with the classic community discovery algorithm CPM, the network function modular analysis algorithm MCODE and the hierarchical network graph structure analysis algorithm BGLL. We analyze the results according to NMI, ARI and F-Measure. Finally, a case study of real data was conducted to ensure the correctness and effectiveness of the algorithm, and to obtain the potential drug combination in the medical prescription.

Download Full-text