Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949200
Xiansheng Rao, Jingjing Song, Xibei Yang, Keyu Liu, Pingxin Wang
One typical case of label noise indicates that some samples have been incorrectly labeled in data. Label noise of training samples will significantly affect the learning performances such that the classification accuracy will be reduced. Presently, many results of identifying samples of incorrect labels have been proposed. Most of them are based on the consideration of classifier based accuracy. Therefore, the performance of used classifier is directly related to the result of filtering samples with noise label. In this paper, a neighborhood strategy is introduced into analyzing label noise data, it is mainly because such classifier is superior to several popular classifiers. Not only the neighborhood classifier based algorithm is designed to remove samples with noise label, but also such type of filter is compared with the nearest neighborhood based filter. The experimental results demonstrate that our neighborhood classifier based filter performs well because higher classification accuracy can be achieved. This study suggests new trends for considering neighborhood approach to complex data.
{"title":"Neighborhood Classifier for Label Noise","authors":"Xiansheng Rao, Jingjing Song, Xibei Yang, Keyu Liu, Pingxin Wang","doi":"10.1109/ICMLC48188.2019.8949200","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949200","url":null,"abstract":"One typical case of label noise indicates that some samples have been incorrectly labeled in data. Label noise of training samples will significantly affect the learning performances such that the classification accuracy will be reduced. Presently, many results of identifying samples of incorrect labels have been proposed. Most of them are based on the consideration of classifier based accuracy. Therefore, the performance of used classifier is directly related to the result of filtering samples with noise label. In this paper, a neighborhood strategy is introduced into analyzing label noise data, it is mainly because such classifier is superior to several popular classifiers. Not only the neighborhood classifier based algorithm is designed to remove samples with noise label, but also such type of filter is compared with the nearest neighborhood based filter. The experimental results demonstrate that our neighborhood classifier based filter performs well because higher classification accuracy can be achieved. This study suggests new trends for considering neighborhood approach to complex data.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126141051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949195
Chun-Ming Tsai, T. Shou, Shao-Chi Chen, J. Hsieh
Every two months, the Taiwan Power Company will dispatch staffs to each household to read numbers in electricity meters to calculate and collect electricity bills. However, these electricity meter staff sometimes read the wrong meter numbers and so calculate the wrong electricity bill. A system that automatically detects the digital region in electricity meter, could reduce this misreading of numbers and calculate the electricity bill correctly, thereby increasing work efficiency. Herein, the deep learning model SSD (Single Shot MultiBox Detector) is applied and fine-turned to detect the digital region in electricity meter to help the Taiwan Power Company staff. From the experimental results, it is demonstrated that the presented deep learning methods detect the digital region better than the pre-trained SSD model. In the testing experiments, the accuracies of the digital region detection are 100% for both our collected data's and fine-tuned SSD, respectively.
{"title":"Use SSD to Detect the Digital Region in Electricity Meter","authors":"Chun-Ming Tsai, T. Shou, Shao-Chi Chen, J. Hsieh","doi":"10.1109/ICMLC48188.2019.8949195","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949195","url":null,"abstract":"Every two months, the Taiwan Power Company will dispatch staffs to each household to read numbers in electricity meters to calculate and collect electricity bills. However, these electricity meter staff sometimes read the wrong meter numbers and so calculate the wrong electricity bill. A system that automatically detects the digital region in electricity meter, could reduce this misreading of numbers and calculate the electricity bill correctly, thereby increasing work efficiency. Herein, the deep learning model SSD (Single Shot MultiBox Detector) is applied and fine-turned to detect the digital region in electricity meter to help the Taiwan Power Company staff. From the experimental results, it is demonstrated that the presented deep learning methods detect the digital region better than the pre-trained SSD model. In the testing experiments, the accuracies of the digital region detection are 100% for both our collected data's and fine-tuned SSD, respectively.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127788567","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949218
Yefei Zhang, Zhidong Zhao, Chunwei Guo, Jingzhou Huang, K. Xu
Personal identification based on ECG signals has been a significant challenge. The performance of an ECG authentication system depends significantly on the features extracted and the classifier subsequently applied. Although recently the deep neural networks based approaches featuring adaptive feature extractions and inherent classifications have attracted attention, they usually require a substantial set of training data. Aiming at tackling these issues, this paper presents a convolutional neural network-based transfer learning approach. It includes transferring the big data-trained GoogLeNet model into our identification task, fine-tuning the model using the ‘finetune’ idea, and adding three adaptive layers behind the original feature layer. The proposed approach not only requires a small set of training data, but also obtains great performance.
{"title":"ECG Biometrics Method Based on Convolutional Neural Network and Transfer Learning","authors":"Yefei Zhang, Zhidong Zhao, Chunwei Guo, Jingzhou Huang, K. Xu","doi":"10.1109/ICMLC48188.2019.8949218","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949218","url":null,"abstract":"Personal identification based on ECG signals has been a significant challenge. The performance of an ECG authentication system depends significantly on the features extracted and the classifier subsequently applied. Although recently the deep neural networks based approaches featuring adaptive feature extractions and inherent classifications have attracted attention, they usually require a substantial set of training data. Aiming at tackling these issues, this paper presents a convolutional neural network-based transfer learning approach. It includes transferring the big data-trained GoogLeNet model into our identification task, fine-tuning the model using the ‘finetune’ idea, and adding three adaptive layers behind the original feature layer. The proposed approach not only requires a small set of training data, but also obtains great performance.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130490074","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The decision-theoretic rough set utilizes Bayesian decision to interpret the thresholds of probabilistic rough set model. That provides a novel semantic description for rough regions in the viewpoint of three-way decision theory and has been applied to numerous fields. However, it lacks the ability to deal with lattice-valued information system (LvIS), in which the condition attribute set consists of multiple types of attributes and their domain constitute lattice. Therefore, this study concentrates on the decision-theoretic rough approach in a LvIS. Then, the total decision cost associated with rough regions is addressed and an attribute reduction algorithm will be designed based on minimum decision cost. Finally, a case study on medical diagnosis is conducted to illustrate the decision procedure and attribute reduction approach.
{"title":"A Decision-Theoretic Rough Set Approach to Lattice-Valued Information System","authors":"Jianhang Yu, Hiroshi Morita, Minghao Chen, Weihua Xu","doi":"10.1109/ICMLC48188.2019.8949263","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949263","url":null,"abstract":"The decision-theoretic rough set utilizes Bayesian decision to interpret the thresholds of probabilistic rough set model. That provides a novel semantic description for rough regions in the viewpoint of three-way decision theory and has been applied to numerous fields. However, it lacks the ability to deal with lattice-valued information system (LvIS), in which the condition attribute set consists of multiple types of attributes and their domain constitute lattice. Therefore, this study concentrates on the decision-theoretic rough approach in a LvIS. Then, the total decision cost associated with rough regions is addressed and an attribute reduction algorithm will be designed based on minimum decision cost. Finally, a case study on medical diagnosis is conducted to illustrate the decision procedure and attribute reduction approach.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126340973","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949323
Yingshan Tao, Haoliang Yuan, Chun Sing Lai, L. Lai
With the increase popularity of multi-view data, multi-view learning has attracted vital attentions in pattern recognition as well as machine learning. Most of existing methods apply in traditional single view learning. However, these methods neglect the complementary information among the views. The aim of multi-view is to discover complementary information and enhance the single view learning result. Multi-view is capable of capture incomplete and different types of information from multiple sources. However, multi-views may contain redundant information. Many multi-view methods assume that multi-views are generated from various view-specific generation matrices. This paper proposes the multi-view collaborative representation classification (MVCRC) algorithm which contains the information of different views and the connection of view-to-view. Experimental results conducted on five practical databases are used to confirm the effectiveness of the proposed approach.
{"title":"Multi-View Collaborative Representation Classification","authors":"Yingshan Tao, Haoliang Yuan, Chun Sing Lai, L. Lai","doi":"10.1109/ICMLC48188.2019.8949323","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949323","url":null,"abstract":"With the increase popularity of multi-view data, multi-view learning has attracted vital attentions in pattern recognition as well as machine learning. Most of existing methods apply in traditional single view learning. However, these methods neglect the complementary information among the views. The aim of multi-view is to discover complementary information and enhance the single view learning result. Multi-view is capable of capture incomplete and different types of information from multiple sources. However, multi-views may contain redundant information. Many multi-view methods assume that multi-views are generated from various view-specific generation matrices. This paper proposes the multi-view collaborative representation classification (MVCRC) algorithm which contains the information of different views and the connection of view-to-view. Experimental results conducted on five practical databases are used to confirm the effectiveness of the proposed approach.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128990831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949234
Chun-Wang Lee
This research is based on fuzzy comprehensive evaluation, and lists the fuzzy rule table for designers to control a scooter, in order to affect smoothness in the product design process of electric scooters for the elderly. Step 1: Use questionnaire survey method to understand the factors considered by the designer in designing the electric scooter for the elderly. Step 2: Establish hierarchical analysis and consider the factor weight set in electric scooter design. Step 3: Establish fuzzy hierarchical analysis, and sum up the evaluation result set, as based on the designer's experience. Step 4: Comprehensively consider the influence of all factors and obtain the judgment result. Step 5: List fuzzy rules as an application method to improve the traditional design of electric scooters for the elderly. This study found that the travel speed showed the greatest influence 24.98% on the set of factors affecting smoothness.
{"title":"A Study on Electric Scooters for the Elderly by Applying Fuzzy Theory","authors":"Chun-Wang Lee","doi":"10.1109/ICMLC48188.2019.8949234","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949234","url":null,"abstract":"This research is based on fuzzy comprehensive evaluation, and lists the fuzzy rule table for designers to control a scooter, in order to affect smoothness in the product design process of electric scooters for the elderly. Step 1: Use questionnaire survey method to understand the factors considered by the designer in designing the electric scooter for the elderly. Step 2: Establish hierarchical analysis and consider the factor weight set in electric scooter design. Step 3: Establish fuzzy hierarchical analysis, and sum up the evaluation result set, as based on the designer's experience. Step 4: Comprehensively consider the influence of all factors and obtain the judgment result. Step 5: List fuzzy rules as an application method to improve the traditional design of electric scooters for the elderly. This study found that the travel speed showed the greatest influence 24.98% on the set of factors affecting smoothness.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117116023","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949317
Hsiao-Chi Li, Chang-Yu Cheng, Chia Chou, Chien-Chang Hsu, Meng-Lin Chang, Y. Chiu, J. Chai
Resting-state functional connectivity analyses have revealed a significant effect on the inter-regional interactions in brain. The brain age prediction based on resting-state functional magnetic resonance imaging has been proved as biomarkers to characterize the typical brain development and neuropsychiatric disorders. The brain age prediction model based on functional connectivity measurements derived from resting-state functional magnetic resonance imaging has received a lots of interest in recent years due to its great success in age prediction. However, some of the recent studies rely on experienced neuroscientist experts to select appropriate connectivity features in order to build a robust model for prediction while the others just selected the features based on trial-and-error test. Besides, the subjects used in this studies omitted some subjects that can be divided into two groups with less similarity which may confused the prediction model. In this study, we proposed a multi-class age categories discrimination method with the connectivity features selected via K-means clustering with no prior knowledge provided. The experimental results show that with K-means selected features the proposed model better discriminate multi-class age categories.
{"title":"Multi-Class Brain Age Discrimination Using Machine Learning Algorithm","authors":"Hsiao-Chi Li, Chang-Yu Cheng, Chia Chou, Chien-Chang Hsu, Meng-Lin Chang, Y. Chiu, J. Chai","doi":"10.1109/ICMLC48188.2019.8949317","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949317","url":null,"abstract":"Resting-state functional connectivity analyses have revealed a significant effect on the inter-regional interactions in brain. The brain age prediction based on resting-state functional magnetic resonance imaging has been proved as biomarkers to characterize the typical brain development and neuropsychiatric disorders. The brain age prediction model based on functional connectivity measurements derived from resting-state functional magnetic resonance imaging has received a lots of interest in recent years due to its great success in age prediction. However, some of the recent studies rely on experienced neuroscientist experts to select appropriate connectivity features in order to build a robust model for prediction while the others just selected the features based on trial-and-error test. Besides, the subjects used in this studies omitted some subjects that can be divided into two groups with less similarity which may confused the prediction model. In this study, we proposed a multi-class age categories discrimination method with the connectivity features selected via K-means clustering with no prior knowledge provided. The experimental results show that with K-means selected features the proposed model better discriminate multi-class age categories.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131275666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949168
Nai-Hsiang Chang, Yi-Hsing Chien, H. Chiang, Wei-Yen Wang, C. Hsu
In this paper, a merged convolution neural network (CNN) framework is proposed to automatically avoid obstacles. Although there are many methods for avoiding obstacles, previous methods mostly contain high energy-consuming and high cost. This paper aims to realize an image-based method with a monocular webcam. The experimental results illustrate that the proposed method can effectively avoid obstacles in mobile robot navigation.
{"title":"A Robot Obstacle Avoidance Method Using Merged CNN Framework","authors":"Nai-Hsiang Chang, Yi-Hsing Chien, H. Chiang, Wei-Yen Wang, C. Hsu","doi":"10.1109/ICMLC48188.2019.8949168","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949168","url":null,"abstract":"In this paper, a merged convolution neural network (CNN) framework is proposed to automatically avoid obstacles. Although there are many methods for avoiding obstacles, previous methods mostly contain high energy-consuming and high cost. This paper aims to realize an image-based method with a monocular webcam. The experimental results illustrate that the proposed method can effectively avoid obstacles in mobile robot navigation.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123266588","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Alternative splicing of messenger RNAs (mRNAs) is a common and conserved biological process in eukaryotes. The aberrancy or disruption of different alternative splicing forms may cause alterations of cell functions and result in diseases. It is proposed that alternative splicing may play a critical role in the mechanisms of carcinogenesis. By studying a large dataset in The Cancer Genome Atlas database, a recent study showed that alternative splicing, particularly exon-exclusion, is a powerful prognosis factor in serious subtype of ovarian cancer. However, the characteristics of alternative splicing has not been studied in other subtypes. In this study, we focus on the alternative splicing events, i.e. single exon-inclusion or -exclusion, in clear-cell ovarian cancer subtype. The subtype appears to have particularly high incidence in Asians comparing to Europeans and Americans and tend to be drug resistant. Transcriptomes were obtained from tumors and their paired-normal tissues from five patients. PSI-values, which represent the proportions of alternative splicing events of an exon, were calculated in both tumor and paired-normal tissues. Differences of PSI-values between tumors and paired-normal were examined by a significant test based on Conditional Beta Regression model. In total, we identified ~200 exons covering 52 genes with significant differences between cancer and paired-normal tissue (p < 0.001) including gene ERP29 and PAM which were previously identified in serous subtypes.
{"title":"Identification of Alternative Splicing Characteristic Associated with Clear-Cell Ovarian Cancer from Paired Normal and Tumor Tissues","authors":"Yu-Ting Huang, M. Shiao, Chen-An Tasi, Kuer-Yuan Lan, Chieh-Hsi Lin, Natini Jianawath, Jia-Ming Chang","doi":"10.1109/ICMLC48188.2019.8949316","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949316","url":null,"abstract":"Alternative splicing of messenger RNAs (mRNAs) is a common and conserved biological process in eukaryotes. The aberrancy or disruption of different alternative splicing forms may cause alterations of cell functions and result in diseases. It is proposed that alternative splicing may play a critical role in the mechanisms of carcinogenesis. By studying a large dataset in The Cancer Genome Atlas database, a recent study showed that alternative splicing, particularly exon-exclusion, is a powerful prognosis factor in serious subtype of ovarian cancer. However, the characteristics of alternative splicing has not been studied in other subtypes. In this study, we focus on the alternative splicing events, i.e. single exon-inclusion or -exclusion, in clear-cell ovarian cancer subtype. The subtype appears to have particularly high incidence in Asians comparing to Europeans and Americans and tend to be drug resistant. Transcriptomes were obtained from tumors and their paired-normal tissues from five patients. PSI-values, which represent the proportions of alternative splicing events of an exon, were calculated in both tumor and paired-normal tissues. Differences of PSI-values between tumors and paired-normal were examined by a significant test based on Conditional Beta Regression model. In total, we identified ~200 exons covering 52 genes with significant differences between cancer and paired-normal tissue (p < 0.001) including gene ERP29 and PAM which were previously identified in serous subtypes.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123958664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/ICMLC48188.2019.8949280
Xiaoyue Liu, Hui-Yi Liu, Jie Gao
This paper proposes an inverse kinematics solution based on a two-layer hierarchical cluster model. It divides a human skeleton into blocks to build a two-layer hierarchical cluster model. Based on the relationship between the end position and angle vectors of joints in the BVH format motion capture data as well as to make the end position vectors of joints into clusters with the K-MEANS cluster method, we then make the angle vectors of each joint into clusters with the nearest-neighbor cluster method. Based on that, the inverse kinematics solution is made with the consistency between frames as the constraint condition. The experiment results show that the method is high accuracy, fast solution speed and strong adaptability.
{"title":"An Inverse Kinematic Solution Based on a Two-Layer Hierarchical Cluster Model","authors":"Xiaoyue Liu, Hui-Yi Liu, Jie Gao","doi":"10.1109/ICMLC48188.2019.8949280","DOIUrl":"https://doi.org/10.1109/ICMLC48188.2019.8949280","url":null,"abstract":"This paper proposes an inverse kinematics solution based on a two-layer hierarchical cluster model. It divides a human skeleton into blocks to build a two-layer hierarchical cluster model. Based on the relationship between the end position and angle vectors of joints in the BVH format motion capture data as well as to make the end position vectors of joints into clusters with the K-MEANS cluster method, we then make the angle vectors of each joint into clusters with the nearest-neighbor cluster method. Based on that, the inverse kinematics solution is made with the consistency between frames as the constraint condition. The experiment results show that the method is high accuracy, fast solution speed and strong adaptability.","PeriodicalId":221349,"journal":{"name":"2019 International Conference on Machine Learning and Cybernetics (ICMLC)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117077363","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：