Sentiment analysis aims to observe and summarize a person's opinions or emotional states through textual data. Despite the demands of sentiment analysis methods for analyzing social media data, fundamental challenges still remained because user-generated data is unstructured, unlabeled, and "noisy". The morphological sentence pattern (MSP) model, an aspect-based lexicon building method, is proposed for dealing with the problems of the transitional sentiment analysis by recognizing the "aspect-expression" in a sentence. However, there are limitations on this model. Firstly, since the MSP model is based on the pattern matching, the sentences cannot be analyzed when the pattern does not exist in the lexicon. Secondly, the patterns should be continuously updated to maintain a high level of accuracy. In this paper, to compensate for the limitations of the MSP model, we proposed Deep-MSP, a deep learning approach based on multiple convolutional neural networks (ConvNet or CNN), designed to recognize whether or not the target part-of-speech has potential to be the aspect-expression from not only existing patterns but also new patterns.
{"title":"Deep-MSP: Morphological Sentence Pattern Recognition based on ConvNet","authors":"S. Park, Youngsub Han, Yanggon Kim","doi":"10.1145/3129676.3129702","DOIUrl":"https://doi.org/10.1145/3129676.3129702","url":null,"abstract":"Sentiment analysis aims to observe and summarize a person's opinions or emotional states through textual data. Despite the demands of sentiment analysis methods for analyzing social media data, fundamental challenges still remained because user-generated data is unstructured, unlabeled, and \"noisy\". The morphological sentence pattern (MSP) model, an aspect-based lexicon building method, is proposed for dealing with the problems of the transitional sentiment analysis by recognizing the \"aspect-expression\" in a sentence. However, there are limitations on this model. Firstly, since the MSP model is based on the pattern matching, the sentences cannot be analyzed when the pattern does not exist in the lexicon. Secondly, the patterns should be continuously updated to maintain a high level of accuracy. In this paper, to compensate for the limitations of the MSP model, we proposed Deep-MSP, a deep learning approach based on multiple convolutional neural networks (ConvNet or CNN), designed to recognize whether or not the target part-of-speech has potential to be the aspect-expression from not only existing patterns but also new patterns.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128238788","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}
Recently, Ransomware has been rapidly increasing and is becoming far more dangerous than other common malware types. Unlike previous versions of Ransomware that infect email attachments or access certain sites, the new Ransomware, such as WannaCryptor, corrupts data even when the PC is connected to the Internet. Therefore, many studies are being conducted to detect and defend Ransomware. However, existing studies on Ransomware detection cannot effectively detect and defend the new Ransomware because it detects Ransomware using signature databases or monitoring specific activities of processes. In this paper, we propose a method to make decoy files for detecting Ransomwares efficiently. The proposed method is based on the analysis of the behaviors of existing Ransomwares at the source code level.
{"title":"How to Make Efficient Decoy Files for Ransomware Detection?","authors":"Jong-Seop Lee, Jinwoo Lee, Jiman Hong","doi":"10.1145/3129676.3129713","DOIUrl":"https://doi.org/10.1145/3129676.3129713","url":null,"abstract":"Recently, Ransomware has been rapidly increasing and is becoming far more dangerous than other common malware types. Unlike previous versions of Ransomware that infect email attachments or access certain sites, the new Ransomware, such as WannaCryptor, corrupts data even when the PC is connected to the Internet. Therefore, many studies are being conducted to detect and defend Ransomware. However, existing studies on Ransomware detection cannot effectively detect and defend the new Ransomware because it detects Ransomware using signature databases or monitoring specific activities of processes. In this paper, we propose a method to make decoy files for detecting Ransomwares efficiently. The proposed method is based on the analysis of the behaviors of existing Ransomwares at the source code level.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128415727","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}
Recognition of handwritten Arabic text is a difficult task since there are many challenges and obstacles that face any handwritten Arabic OCR system. Some of them include, but are not limited to: different handwriting styles, different characters that have similar contours, and the same character may have different forms according to its position in a sentence. Several approaches have been attempted to accurately recognize handwritten Arabic characters. However, the issue of the accuracy of Arabic OCR in handwritten text continues to be a dilemma. We will describe the general difficulties in handwritten Arabic language text, and propose a novel approach for identifying isolated handwritten Arabic characters using encoded Freeman chain code. We will also apply a novel approach of using change in tangents to classify characters. Several handwritten Arabic characters were trained and tested with our own dataset. The results showed the efficacy of our approach for recognizing isolated handwritten Arabic characters. The average accuracy rate of our method ranges from 92% to 97%.
{"title":"Isolated Handwritten Arabic Character Recognition Using Freeman Chain Code and Tangent Line","authors":"Hassan Althobaiti, Kevat Shah, Chao Lu","doi":"10.1145/3129676.3129678","DOIUrl":"https://doi.org/10.1145/3129676.3129678","url":null,"abstract":"Recognition of handwritten Arabic text is a difficult task since there are many challenges and obstacles that face any handwritten Arabic OCR system. Some of them include, but are not limited to: different handwriting styles, different characters that have similar contours, and the same character may have different forms according to its position in a sentence. Several approaches have been attempted to accurately recognize handwritten Arabic characters. However, the issue of the accuracy of Arabic OCR in handwritten text continues to be a dilemma. We will describe the general difficulties in handwritten Arabic language text, and propose a novel approach for identifying isolated handwritten Arabic characters using encoded Freeman chain code. We will also apply a novel approach of using change in tangents to classify characters. Several handwritten Arabic characters were trained and tested with our own dataset. The results showed the efficacy of our approach for recognizing isolated handwritten Arabic characters. The average accuracy rate of our method ranges from 92% to 97%.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127994239","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}
With thousands of malware samples pouring out every day, how can we reduce malware analysis time and detect them effectively? Malware family classification provides one of good measures to predict characteristics of unknown malware since malware belonging to the same family can have similar features. Static analysis and dynamic analysis are techniques to obtain features to be used for classifying malware samples to their families. Static analysis performs analysis based on specific signatures included in the malware. Static analysis has the advantages that the scope of the analysis covers the entire code, and the analysis can be performed without executing the malware. However, it is very difficult to detect or classify malware variants with only the results of the static analysis, because malware developers use polymorphic or encryption techniques to avoid static analysis-based detection of anti-virus software. Dynamic analysis analyzes malware behaviors, so the results of dynamic analysis can be used to detect or classify malware variants. One of dynamic features that can be used to detect or classify malware variants is API call sequences. In this paper, we propose a novel method to extract representative API call patterns of malware families using Recurrent Neural Network (RNN). We conducted experiments with 787 malware samples belonging to 9 families. In our experiments, we extracted representative API call patterns of 9 malware families on 551 samples as a training set and performed classification on the 236 samples as a test set. Classification accuracy results using API call patterns extracted from RNN were measured as 71% on average. The results show the feasibility of our approach using RNN to extract representative API call pattern of malware families for malware family classification.
{"title":"Extracting the Representative API Call Patterns of Malware Families Using Recurrent Neural Network","authors":"Iltaek Kwon, E. Im","doi":"10.1145/3129676.3129712","DOIUrl":"https://doi.org/10.1145/3129676.3129712","url":null,"abstract":"With thousands of malware samples pouring out every day, how can we reduce malware analysis time and detect them effectively? Malware family classification provides one of good measures to predict characteristics of unknown malware since malware belonging to the same family can have similar features. Static analysis and dynamic analysis are techniques to obtain features to be used for classifying malware samples to their families. Static analysis performs analysis based on specific signatures included in the malware. Static analysis has the advantages that the scope of the analysis covers the entire code, and the analysis can be performed without executing the malware. However, it is very difficult to detect or classify malware variants with only the results of the static analysis, because malware developers use polymorphic or encryption techniques to avoid static analysis-based detection of anti-virus software. Dynamic analysis analyzes malware behaviors, so the results of dynamic analysis can be used to detect or classify malware variants. One of dynamic features that can be used to detect or classify malware variants is API call sequences. In this paper, we propose a novel method to extract representative API call patterns of malware families using Recurrent Neural Network (RNN). We conducted experiments with 787 malware samples belonging to 9 families. In our experiments, we extracted representative API call patterns of 9 malware families on 551 samples as a training set and performed classification on the 236 samples as a test set. Classification accuracy results using API call patterns extracted from RNN were measured as 71% on average. The results show the feasibility of our approach using RNN to extract representative API call pattern of malware families for malware family classification.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132871836","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}
Appearance of microcalcifications in mammograms is one of the early signs of breast cancer. In this work, one-class support vector machines (SVM), a novelty detection method, is utilized for detection of the mammogram samples containing microcalcifications. These samples are small regions of the mammograms with the size of 25x25 pixels. Each of the samples are represented by 25 features that are already proven to be accurate identifiers of the microcalcifications. Since the obtained classification performance of one-class SVM with all these 25 features is very low (accuracy = 0.5575, sensitivity = 0.2107, specificity = 0.9042), number of these features is reduced by using principal component analysis (PCA). Training a classifier only with the PCA features achieves an improved performance (accuracy = 0.9464, sensitivity = 1.0000, specificity = 0.8927) where the number of false negative samples is reduced from 206 to 0.
{"title":"A Novelty Detection Approach to Classification of Breast Tissue Containing Microcalcifications","authors":"E. Avşar, Kurtuluş Buluş","doi":"10.1145/3129676.3129680","DOIUrl":"https://doi.org/10.1145/3129676.3129680","url":null,"abstract":"Appearance of microcalcifications in mammograms is one of the early signs of breast cancer. In this work, one-class support vector machines (SVM), a novelty detection method, is utilized for detection of the mammogram samples containing microcalcifications. These samples are small regions of the mammograms with the size of 25x25 pixels. Each of the samples are represented by 25 features that are already proven to be accurate identifiers of the microcalcifications. Since the obtained classification performance of one-class SVM with all these 25 features is very low (accuracy = 0.5575, sensitivity = 0.2107, specificity = 0.9042), number of these features is reduced by using principal component analysis (PCA). Training a classifier only with the PCA features achieves an improved performance (accuracy = 0.9464, sensitivity = 1.0000, specificity = 0.8927) where the number of false negative samples is reduced from 206 to 0.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115039761","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 strategic application domains of Cyber Physical Systems (CPS) include health care, transportation, management of large-scale physical infrastructures, and power plants. In all these applications the systems need to adapt, depending upon the availability of reliable resources, in order to provide trustworthy services at every context of its execution. Hence, CPS is viewed as a large distributed system of service and supply chain management, in which services are resource-centric. For service adaptation, resource quality and availability are determining factors, especially during emergencies. In this paper an abstract service-oriented description of CPS is given with emphasis on resource and service providers and requesters, the context space created by them is defined, and adaptation rules that arise from the context constraints and contracts defined by them are explained.
{"title":"Modeling Resource-centric Services for Service Adaptation in Cyber Physical Systems","authors":"V. Alagar, Kaiyu Wan","doi":"10.1145/3129676.3129694","DOIUrl":"https://doi.org/10.1145/3129676.3129694","url":null,"abstract":"The strategic application domains of Cyber Physical Systems (CPS) include health care, transportation, management of large-scale physical infrastructures, and power plants. In all these applications the systems need to adapt, depending upon the availability of reliable resources, in order to provide trustworthy services at every context of its execution. Hence, CPS is viewed as a large distributed system of service and supply chain management, in which services are resource-centric. For service adaptation, resource quality and availability are determining factors, especially during emergencies. In this paper an abstract service-oriented description of CPS is given with emphasis on resource and service providers and requesters, the context space created by them is defined, and adaptation rules that arise from the context constraints and contracts defined by them are explained.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"349 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122839675","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}
This study aims at developing and applying a framework that supports IPMI(Intelligent Platform Management Interface) and DCMI(Data Center Manageability Interface) based on open BMC(Board Support Controller). The proposed framework includes IPMI 2.0-based control firmware and DCMI technology that enables server board management in the data center. The existing BMC has been provided in the form of a private binary by a third-party vendor, and needs to be consulted with vendors when changes such as addition / deletion are required. In addition, the functionality provided by BMC has been determined by vendor-specific characteristics. In this paper, the proposed open BMC provides BMC IDE environment as well as code release. Therefore, user (developer) can configure BMC function more easily, and it is designed to implement and delete additional function if necessary. The designed BMC IDE environment not only supports BMC Application development, but also helps board system providers to design and evaluate IPMI library and DCMI library.
本研究旨在开发和应用一个基于开放BMC(Board Support Controller)的支持IPMI(Intelligent Platform Management Interface)和DCMI(Data Center Manageability Interface)的框架。提出的框架包括基于IPMI 2.0的控制固件和支持数据中心服务器板管理的DCMI技术。现有的BMC已由第三方供应商以私有二进制文件的形式提供，当需要进行诸如添加/删除之类的更改时，需要与供应商协商。此外，BMC提供的功能是由特定于供应商的特性决定的。本文提出的开放式BMC提供BMC IDE环境和代码发布。因此，用户(开发人员)可以更轻松地配置BMC功能，并且可以根据需要实现和删除附加功能。所设计的BMC IDE环境不仅支持BMC应用程序的开发，还可以帮助板系统供应商设计和评估IPMI库和DCMI库。
{"title":"Design of Framework supporting IPMI and DCMI based on Open BMC","authors":"J. An, Younghwan Kim, Changkwon Park","doi":"10.1145/3129676.3129706","DOIUrl":"https://doi.org/10.1145/3129676.3129706","url":null,"abstract":"This study aims at developing and applying a framework that supports IPMI(Intelligent Platform Management Interface) and DCMI(Data Center Manageability Interface) based on open BMC(Board Support Controller). The proposed framework includes IPMI 2.0-based control firmware and DCMI technology that enables server board management in the data center. The existing BMC has been provided in the form of a private binary by a third-party vendor, and needs to be consulted with vendors when changes such as addition / deletion are required. In addition, the functionality provided by BMC has been determined by vendor-specific characteristics. In this paper, the proposed open BMC provides BMC IDE environment as well as code release. Therefore, user (developer) can configure BMC function more easily, and it is designed to implement and delete additional function if necessary. The designed BMC IDE environment not only supports BMC Application development, but also helps board system providers to design and evaluate IPMI library and DCMI library.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122285860","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}
Dong-Kyu Chae, B. Kim, Seung-Ho Kim, Sang-Wook Kim
In this paper, we introduce a novel problem of dynamic graph bag classification, and propose a method to solve this problem. Here, a graph bag (simply, bag) corresponds to a training object that contains one or multiple graphs. Dynamic bag classification aims to build a classification model for bags which are presented in a dynamic fashion, i.e., emerging of new bags or graphs. Our proposed solution for this problem can gradually update the classification model whenever such changes are made to a bag dataset, rather than building a model from the scratch. We demonstrate the effectiveness of our proposed method by our extensive evaluation on a real-world graph dataset.
{"title":"On Classifying Dynamic Graph Bags","authors":"Dong-Kyu Chae, B. Kim, Seung-Ho Kim, Sang-Wook Kim","doi":"10.1145/3129676.3129730","DOIUrl":"https://doi.org/10.1145/3129676.3129730","url":null,"abstract":"In this paper, we introduce a novel problem of dynamic graph bag classification, and propose a method to solve this problem. Here, a graph bag (simply, bag) corresponds to a training object that contains one or multiple graphs. Dynamic bag classification aims to build a classification model for bags which are presented in a dynamic fashion, i.e., emerging of new bags or graphs. Our proposed solution for this problem can gradually update the classification model whenever such changes are made to a bag dataset, rather than building a model from the scratch. We demonstrate the effectiveness of our proposed method by our extensive evaluation on a real-world graph dataset.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129736937","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}
Jinmang Jung, Jisu Park, Seoyeon Kim, Mhanwoo Heo, Jiman Hong
Paravirtualization manages virtual machines and virtual resources efficiently by the communication between the virtualization layer and modified guest OSes. In a paravirtual environment, the I/O response of a virtual machine is hard to approach that of a native OS because a virtual I/O is asynchronously processed by the virtualization layer without hardware supports. Virtual CPU scheduling algorithms have been proposed to improve the I/O performance. However, existing solutions lack the I/O fairness when virtual machines have various or skewed of workloads because they put the I/O performance of latency-sensitive vCPUs before vCPUs that are not. In this paper, we design a credit based vCPU scheduling model for I/O performance of virtual machines by using a loan and repayment system. Credit rating of each virtual CPU is periodically evaluated by observing its resource consumption pattern and a virtual CPU cannot be allocated more resources until the repayment is finished.
{"title":"A Virtual CPU Scheduling Model for I/O Performance in Paravirtualized Environments","authors":"Jinmang Jung, Jisu Park, Seoyeon Kim, Mhanwoo Heo, Jiman Hong","doi":"10.1145/3129676.3131703","DOIUrl":"https://doi.org/10.1145/3129676.3131703","url":null,"abstract":"Paravirtualization manages virtual machines and virtual resources efficiently by the communication between the virtualization layer and modified guest OSes. In a paravirtual environment, the I/O response of a virtual machine is hard to approach that of a native OS because a virtual I/O is asynchronously processed by the virtualization layer without hardware supports. Virtual CPU scheduling algorithms have been proposed to improve the I/O performance. However, existing solutions lack the I/O fairness when virtual machines have various or skewed of workloads because they put the I/O performance of latency-sensitive vCPUs before vCPUs that are not. In this paper, we design a credit based vCPU scheduling model for I/O performance of virtual machines by using a loan and repayment system. Credit rating of each virtual CPU is periodically evaluated by observing its resource consumption pattern and a virtual CPU cannot be allocated more resources until the repayment is finished.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132247273","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 research interest of real-time global illumination has increased due to the growing demand of graphics applications such as virtual reality. Recently, the design that combines Image-based rendering (IBR) and Ray-Tracing to create Synthetic Light Field (SLF) has been widely adopted to provide delicate visual experience for multiple viewpoints at an acceptable frame rate. However, despite its parallel characteristic, constructing a SLF is still inefficient on modern Graphics Processing Unit (GPU) due to the irregularities. For instance, the issues caused by branch divergence, early-termination and irregular memory access prolong the execution time that cannot be simply resolved by workload merging. In this paper, we proposed a Runtime design that reorganizes the execution into a pipeline-based pattern with grouping of primary rays. With this approach, the number of valid rays can be maintained at a high level with less divergence of paths. Based on the experiment on a heterogeneous system, the throughput becomes 2.48 times higher than the original on average.
{"title":"Efficient Synthetic Light Field Rendering on Heterogeneous Systems Using a Pipeline-Based Runtime Design","authors":"Chih-Chen Kao, Liang-Chi Tseng, W. Hsu","doi":"10.1145/3129676.3129677","DOIUrl":"https://doi.org/10.1145/3129676.3129677","url":null,"abstract":"The research interest of real-time global illumination has increased due to the growing demand of graphics applications such as virtual reality. Recently, the design that combines Image-based rendering (IBR) and Ray-Tracing to create Synthetic Light Field (SLF) has been widely adopted to provide delicate visual experience for multiple viewpoints at an acceptable frame rate. However, despite its parallel characteristic, constructing a SLF is still inefficient on modern Graphics Processing Unit (GPU) due to the irregularities. For instance, the issues caused by branch divergence, early-termination and irregular memory access prolong the execution time that cannot be simply resolved by workload merging. In this paper, we proposed a Runtime design that reorganizes the execution into a pipeline-based pattern with grouping of primary rays. With this approach, the number of valid rays can be maintained at a high level with less divergence of paths. Based on the experiment on a heterogeneous system, the throughput becomes 2.48 times higher than the original on average.","PeriodicalId":326100,"journal":{"name":"Proceedings of the International Conference on Research in Adaptive and Convergent Systems","volume":"31 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113976383","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：