Proceedings of the 3rd International Conference on Machine Learning and Soft Computing最新文献_第3页

Real-Time Multi-Camera Multi-Person Action Recognition using Pose Estimation 基于姿态估计的实时多摄像头多人动作识别

Proceedings of the 3rd International Conference on Machine Learning and Soft Computing Pub Date : 2019-01-25 DOI: 10.1145/3310986.3311006

Jonathan Then Sien Phang, K. Lim

{"title":"Real-Time Multi-Camera Multi-Person Action Recognition using Pose Estimation","authors":"Jonathan Then Sien Phang, K. Lim","doi":"10.1145/3310986.3311006","DOIUrl":"https://doi.org/10.1145/3310986.3311006","url":null,"abstract":"Action recognition possesses challenging issues in real-time multi-camera scenario when dealing with multi-person such as occlusion, pose variance and action interaction. In this paper, a real-time pipeline is proposed to address multi-person action recognition in multi-camera setup using joint key-points sequences from detected person. Joints trajectory is the important time-series information to identify actions. 14 key-points from human joints are scaled with relative to the Euclidean distance of neck-to-pelvis to obtain standard size of person, which is invariant to camera distance. Subsequently, 3D histogram correlation is applied to match multi-person identity. An indexed person with a series of action attribute are collected and fed into Long Short-Term Memory (LSTM) recurrent neural network. The proposed pipeline uses spatial-temporal feature of person's joint key-points trajectory for action recognition. Minimal single pass forward time through the LSTM network enables real-time multi-person action recognition in a video sequence. The proposed pipeline achieved up to 13 frames per second with 92% recognition rate with two camera setups.","PeriodicalId":252781,"journal":{"name":"Proceedings of the 3rd International Conference on Machine Learning and Soft Computing","volume":"52 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113961201","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}

引用次数: 5

Application of Improved Differential Evolution Algorithm for Economic and Emission Dispatch of Thermal Power Generation Plants 改进差分进化算法在火力发电厂经济排放调度中的应用

Proceedings of the 3rd International Conference on Machine Learning and Soft Computing Pub Date : 2019-01-25 DOI: 10.1145/3310986.3311003

Nguyen Tan Hung, Nguyen Hung, P. Nguyen, Dinh Thanh Viet

引用次数: 1

Multi-view Neural Network Integrating Knowledge for Patient Self-diagnosis 集成知识的多视图神经网络用于患者自我诊断

Proceedings of the 3rd International Conference on Machine Learning and Soft Computing Pub Date : 2019-01-25 DOI: 10.1145/3310986.3311016

Fangyuan Zhao, Jianliang Xu, Yong Lin

引用次数: 0

A System Analysis and Design of Marketing Strategy for Improving Pineapple Agritourism 菠萝农业发展营销策略的系统分析与设计

Proceedings of the 3rd International Conference on Machine Learning and Soft Computing Pub Date : 2019-01-25 DOI: 10.1145/3310986.3311009

Kartika Trianita, Taufik Djatna, A. Fauzi

{"title":"A System Analysis and Design of Marketing Strategy for Improving Pineapple Agritourism","authors":"Kartika Trianita, Taufik Djatna, A. Fauzi","doi":"10.1145/3310986.3311009","DOIUrl":"https://doi.org/10.1145/3310986.3311009","url":null,"abstract":"Marketing strategy is the most important factor to increase the visitor's number to an agritourism. Accordingly, the selection of the best marketing strategy is needed so it provides the maximum result. Failure of strategy implementation can cause income loss for agritourism operational. Hence, adequate system analysis and business process design are needed to reduce the possibility of failure. This paper aims to analyze the requirements of the system and to design the business process including identify and determine visitor preferences in agritourism facilities and determine association rules of the priority marketing strategy for improving pineapple agritourism. Analytical system entity construction was used to describe the requirements of the system, the Unified Modeling Language and Business Process Model and Notation 2.0 were used to design the business process. Due to the limited marketing budget, the top eight most interesting facilities based on visitor preferences were determined by RELIEF-F algorithm. Then, the actionable marketing rules were determined by ARM algorithm. The result of system analysis shows that the system requires facilities and visitor preferences as inputs, agritourism operational and visitor as stakeholders, to result in an output of association rules as improved marketing strategies. From the design, we obtained the top eight facilities are camping ground, photo spot, pineapple field tour, pineapple factory tour, culture attraction, bicycle track, gazebo, and playground. ARM results association rules design is composed of five facilities: pineapple factory, pineapple field, gazebo, playground, and photo spot. From the result, we can conclude that the agritourism operational should consider placing the facilities close to each other according to obtained association rules.","PeriodicalId":252781,"journal":{"name":"Proceedings of the 3rd International Conference on Machine Learning and Soft Computing","volume":"314 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127565892","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}

引用次数: 1

A New Mammography Lesion Classification Method Based on Convolutional Neural Network 一种基于卷积神经网络的乳腺x线造影病灶分类新方法

Proceedings of the 3rd International Conference on Machine Learning and Soft Computing Pub Date : 2019-01-25 DOI: 10.1145/3310986.3311019

Xinlei Wei, Yide Ma, Runze Wang

引用次数: 1

A method for Representation the Knowledge of Functions and Operators and Application 函数和运算符知识的表示方法及其应用

Proceedings of the 3rd International Conference on Machine Learning and Soft Computing Pub Date : 2019-01-25 DOI: 10.1145/3310986.3311012

H. Nguyen, Son T. Luu, N. Do

引用次数: 2

Padding Methods in Convolutional Sequence Model: An Application in Japanese Handwriting Recognition 卷积序列模型填充方法在日文手写识别中的应用

Proceedings of the 3rd International Conference on Machine Learning and Soft Computing Pub Date : 2019-01-25 DOI: 10.1145/3310986.3310998

Nguyen Tuan Nam, P. D. Hung

{"title":"Padding Methods in Convolutional Sequence Model: An Application in Japanese Handwriting Recognition","authors":"Nguyen Tuan Nam, P. D. Hung","doi":"10.1145/3310986.3310998","DOIUrl":"https://doi.org/10.1145/3310986.3310998","url":null,"abstract":"Today, there is a wide range of research cases about end-to-end trained and sequence-to-sequence models applied in the task of handwritten character recognition. Most of which mark the combination between convolutional neural network (CNN) as a feature extraction module and recurrent neural network (RNN) as a sequence-to-sequence module. Notably, the CNN layer can be fed with dynamic sizes of input images while the RNN layer can tolerate dynamic lengths of input data, which subsequently makes up the dynamic feature of the recognition models. However, when the number one priority is to minimize the training timespan, the models are to receive training data in the form of mini-batch, which requires resizing or padding images into an equal size instead of using original multiple-size pictures due to the fact that most of the deep learning frameworks (such as keras, tensorflow, caffe, etc.) only accept the same-size input and output in one mini-batch. Actually, this practice may lower the model dynamicity in the training process. So, the question is whether it might be a trade-off between the effectiveness (level of accuracy) and the time optimization of the model. In this paper, we will examine different impact of various padding and non-padding methods on the same model architecture for Japanese handwriting recognition before finally concluding on which method has the most reasonable training time but can produce an accuracy rate of up to 95%.","PeriodicalId":252781,"journal":{"name":"Proceedings of the 3rd International Conference on Machine Learning and Soft Computing","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115366161","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}

引用次数: 23

Geographic Entity Relationship Extraction Model Based on Piecewise Convolution of Residual Network 基于残差网络分段卷积的地理实体关系提取模型

Proceedings of the 3rd International Conference on Machine Learning and Soft Computing Pub Date : 2019-01-25 DOI: 10.1145/3310986.3311025

Ying Jin, Shuai Zhao, Yudong Wu

引用次数: 1

Parasite worm egg automatic detection in microscopy stool image based on Faster R-CNN 基于Faster R-CNN的显微镜粪便图像中寄生虫虫卵自动检测

Proceedings of the 3rd International Conference on Machine Learning and Soft Computing Pub Date : 2019-01-25 DOI: 10.1145/3310986.3311014

Ngo Quoc Viet, Dang Thi ThanhTuyen, Trinh Huy Hoang

引用次数: 14

A Developing Method for Distributed Sensing Systems 分布式传感系统的开发方法

Proceedings of the 3rd International Conference on Machine Learning and Soft Computing Pub Date : 2019-01-25 DOI: 10.1145/3310986.3311020

T. Truong, B. Pottier, H. Huynh

引用次数: 5