The generalization of knowledge is a necessary part of every scientific field. Meta-analysis is already advocated as a tool for generalization in different areas such as medicine, psychology, business, and this process is already standardized for them. Software engineering started using meta-analysis as a tool for aggregating results from families of experiments, but not so long for generalization of results coming from different studies, and for this purpose, the meta-analytical approach is not yet clarified. In this paper, we attempt to systematize the application of meta-analysis as a secondary study to the software engineering field suggesting our preliminary protocol. To see the reliability of the proposed protocol we conducted several studies using it. Following even uniform protocol with these studies, we identified the issues preventing the wide usage of meta-analysis in software engineering and proposed our solutions for them.
{"title":"Systematizing the Meta-Analytical Process in Software Engineering","authors":"M. L. Jaccheri, Zamira Kholmatova, G. Succi","doi":"10.1145/3501774.3501775","DOIUrl":"https://doi.org/10.1145/3501774.3501775","url":null,"abstract":"The generalization of knowledge is a necessary part of every scientific field. Meta-analysis is already advocated as a tool for generalization in different areas such as medicine, psychology, business, and this process is already standardized for them. Software engineering started using meta-analysis as a tool for aggregating results from families of experiments, but not so long for generalization of results coming from different studies, and for this purpose, the meta-analytical approach is not yet clarified. In this paper, we attempt to systematize the application of meta-analysis as a secondary study to the software engineering field suggesting our preliminary protocol. To see the reliability of the proposed protocol we conducted several studies using it. Following even uniform protocol with these studies, we identified the issues preventing the wide usage of meta-analysis in software engineering and proposed our solutions for them.","PeriodicalId":255059,"journal":{"name":"Proceedings of the 2021 European Symposium on Software Engineering","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122549777","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}
In this paper we present the method and tool for linking the analysis of a software at the development stage with the efficiency of the developed product during operation mode from the energy consumption perspective. The purpose of the method is to recognize the bottlenecks of a program and provide recommendations for improving the structure and run-time behavior of the software. The developed tool consists of two subsystems: first is responsible for static analysis of the code and relevant software metrics, second performs analysis of the power consumption of the application. Analysis of the outputs of both components allows us to create a close-loop system for continuous analysis and optimization of the developing software product.
{"title":"Incorporating energy efficiency measurement into CICD pipeline","authors":"A. Kruglov, G. Succi, X. Vasquez","doi":"10.1145/3501774.3501777","DOIUrl":"https://doi.org/10.1145/3501774.3501777","url":null,"abstract":"In this paper we present the method and tool for linking the analysis of a software at the development stage with the efficiency of the developed product during operation mode from the energy consumption perspective. The purpose of the method is to recognize the bottlenecks of a program and provide recommendations for improving the structure and run-time behavior of the software. The developed tool consists of two subsystems: first is responsible for static analysis of the code and relevant software metrics, second performs analysis of the power consumption of the application. Analysis of the outputs of both components allows us to create a close-loop system for continuous analysis and optimization of the developing software product.","PeriodicalId":255059,"journal":{"name":"Proceedings of the 2021 European Symposium on Software Engineering","volume":" 21","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120828824","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}
In this paper, we introduce a variety of acquisition functions currently used in Bayesian optimization. Besides the traditional acquisition functions like Probability Improvement (PI), Expected Improvement (EI) and Gaussian Process-Upper Confidence Bound (GP-UCB), we also present some modified or improved EI and PI methods, Knowledge Gradient (KG) and Predictive Entropy Search (PES) methods to explore ways to reduce the impact of observational noise. In experimental part, we choose a benchmark function and use Bayesian optimization algorithm to find its global minimum. We add different scales of noise in particular following the Gaussian distribution to the benchmark function, to compare the performance of BO algorithm using different acquisition functions. Combined with the experimental results, we also present a discussion of the pros and cons of using those acquisition functions. Hope this can provide some experience and suggestions for choosing acquisition functions in terms of noisy observation.
{"title":"Alternative Acquisition Functions of Bayesian Optimization in terms of Noisy Observation","authors":"Jia-yi Hu, Yuze Jiang, Jiayu Li, Tianyue Yuan","doi":"10.1145/3501774.3501791","DOIUrl":"https://doi.org/10.1145/3501774.3501791","url":null,"abstract":"In this paper, we introduce a variety of acquisition functions currently used in Bayesian optimization. Besides the traditional acquisition functions like Probability Improvement (PI), Expected Improvement (EI) and Gaussian Process-Upper Confidence Bound (GP-UCB), we also present some modified or improved EI and PI methods, Knowledge Gradient (KG) and Predictive Entropy Search (PES) methods to explore ways to reduce the impact of observational noise. In experimental part, we choose a benchmark function and use Bayesian optimization algorithm to find its global minimum. We add different scales of noise in particular following the Gaussian distribution to the benchmark function, to compare the performance of BO algorithm using different acquisition functions. Combined with the experimental results, we also present a discussion of the pros and cons of using those acquisition functions. Hope this can provide some experience and suggestions for choosing acquisition functions in terms of noisy observation.","PeriodicalId":255059,"journal":{"name":"Proceedings of the 2021 European Symposium on Software Engineering","volume":"265 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131735315","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}
Aleksandr N. Nebaba, I. Savvas, M. Butakova, A. Chernov, Petr S. Shevchuk
Machine learning approaches and algorithms are spreading in wide areas in research and technology. Cybersecurity breaches are the common anomalies for networked and distributed infrastructures which are monitored, registered, and described carefully. However, the description of each security breaches episode and its classification is still a difficult problem, especially in highly complex telecommunication infrastructure. Railway information infrastructure usually has a large scale and large diversity of possible security breaches. Today's situation shows the registering of the security breaches has a mature and stable character, but the problem of their automated classification is not solved completely. Many studies on security breaches multiclass classification show inadequate accuracy of classification. We investigated the origins of this problem and suggested the possible roots consist in disbalance the datasets used for machine learning multiclass classification. Thus, we proposed an approach to improve the accuracy of the classification and verified our approach on the really collected datasets with cybersecurity breaches in railway telecommunication infrastructure. We analyzed the results of applying three imbalanced learning methodologies, namely random oversampling, synthetic minority oversampling technique, and the last one with Tomek links. We have implemented three machine learning algorithms, namely Naïve Bayes, K-means, and support vector machine, on disbalances and balanced data to estimate imbalance learning methodologies with comparing results. The proposed approach demonstrated the increase of the accuracy for multiclass classification in the range from 30 to 41%, depending on the imbalanced learning technique.
{"title":"Improving Multiclass Classification of Cybersecurity Breaches in Railway Infrastructure using Imbalanced Learning","authors":"Aleksandr N. Nebaba, I. Savvas, M. Butakova, A. Chernov, Petr S. Shevchuk","doi":"10.1145/3501774.3501789","DOIUrl":"https://doi.org/10.1145/3501774.3501789","url":null,"abstract":"Machine learning approaches and algorithms are spreading in wide areas in research and technology. Cybersecurity breaches are the common anomalies for networked and distributed infrastructures which are monitored, registered, and described carefully. However, the description of each security breaches episode and its classification is still a difficult problem, especially in highly complex telecommunication infrastructure. Railway information infrastructure usually has a large scale and large diversity of possible security breaches. Today's situation shows the registering of the security breaches has a mature and stable character, but the problem of their automated classification is not solved completely. Many studies on security breaches multiclass classification show inadequate accuracy of classification. We investigated the origins of this problem and suggested the possible roots consist in disbalance the datasets used for machine learning multiclass classification. Thus, we proposed an approach to improve the accuracy of the classification and verified our approach on the really collected datasets with cybersecurity breaches in railway telecommunication infrastructure. We analyzed the results of applying three imbalanced learning methodologies, namely random oversampling, synthetic minority oversampling technique, and the last one with Tomek links. We have implemented three machine learning algorithms, namely Naïve Bayes, K-means, and support vector machine, on disbalances and balanced data to estimate imbalance learning methodologies with comparing results. The proposed approach demonstrated the increase of the accuracy for multiclass classification in the range from 30 to 41%, depending on the imbalanced learning technique.","PeriodicalId":255059,"journal":{"name":"Proceedings of the 2021 European Symposium on Software Engineering","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125507902","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}
Spyridon Kourtis, Apostolos Xenakis, K. Kalovrektis, Antonios Plageras, Ioanna Chalvantzi
Digital technologies help students to delve into the process of scientific discovery. Curriculum integration of STEM contents based on constructivism theories of learning as a context to implement the Science, Technology, Engineering, Mathematics (STEM) epistemology. In this work, we proposed an Inquiry based STEM constructivist teaching and learning methodology for Historical events of the Greek Revolution, enriched with smart learning technologies and engineering practices, while combining Educational Robotics approach. In particular, students design and built a history – line and program their robot to move towards each station, which represent an important timestamp event for Greek Revolution. The robot reads the station's QR code, via a smartphone attached to it, and then returns back to its base. Students retrieve information regarding historical events and according to an APP on their smartphone, they play a knowledge game, concerning these events. This work also highlights, via qualitative results, the additional value that smart learning technologiesand educational robotics give to theoretical subjects, such as History, which should need to be taught in an exploratory way. In that way, students are actively involved in hands – on activities and engineering design process, gaining knowledge from subjects as technology, computer science and engineering.
{"title":"An Exploratory Teaching Proposal of Greek History Independence Events based on STEAM Epistemology, Educational Robotics and Smart Learning Technologies","authors":"Spyridon Kourtis, Apostolos Xenakis, K. Kalovrektis, Antonios Plageras, Ioanna Chalvantzi","doi":"10.1145/3501774.3501792","DOIUrl":"https://doi.org/10.1145/3501774.3501792","url":null,"abstract":"Digital technologies help students to delve into the process of scientific discovery. Curriculum integration of STEM contents based on constructivism theories of learning as a context to implement the Science, Technology, Engineering, Mathematics (STEM) epistemology. In this work, we proposed an Inquiry based STEM constructivist teaching and learning methodology for Historical events of the Greek Revolution, enriched with smart learning technologies and engineering practices, while combining Educational Robotics approach. In particular, students design and built a history – line and program their robot to move towards each station, which represent an important timestamp event for Greek Revolution. The robot reads the station's QR code, via a smartphone attached to it, and then returns back to its base. Students retrieve information regarding historical events and according to an APP on their smartphone, they play a knowledge game, concerning these events. This work also highlights, via qualitative results, the additional value that smart learning technologiesand educational robotics give to theoretical subjects, such as History, which should need to be taught in an exploratory way. In that way, students are actively involved in hands – on activities and engineering design process, gaining knowledge from subjects as technology, computer science and engineering.","PeriodicalId":255059,"journal":{"name":"Proceedings of the 2021 European Symposium on Software Engineering","volume":"78 3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116273857","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}
Dimitrios Gravanis, George Kakarontzas, V. Gerogiannis
Within the past decade, the advent of cloud computing in terms of infrastructure, technology stacks, availability of services and tooling, along with the gradual improvement of its market environment, has driven many organizations to either consider or migrate many existing software systems to the cloud, either fully or partially. A common predicament in most cases, is the existence of a complex, monolithic application, potentially considered legacy at the time, that was not designed to be cloud-native and therefore requires a degree of redesign/reimplementation in order to benefit from cloud deployment. In such cases, the decomposition of the monolith to a set of loosely coupled, highly cohesive and self-contained microservices is a valid recommendation, provided that the organization is prepared to withstand the additional cost, in terms of human and financial resources, along with the unavoidable development overhead, which is inevitable during the early stages. However, the tendency of the tech world to embrace new trends and jump on hype trains for fear of obsoletion, has led to an excessive adoption of the microservices architecture (MSA), even in cases where such an architecture is not viable for the organization, or does not derive from any business requirements. This research focuses on establishing the position of a traditional monolith in the modern software architecture landscape and determine use cases that can still benefit from this paradigm, as well as use cases that could benefit from a partial or full transition to microservices architectures instead.
{"title":"You don't need a Microservices Architecture (yet): Monoliths may do the trick","authors":"Dimitrios Gravanis, George Kakarontzas, V. Gerogiannis","doi":"10.1145/3501774.3501780","DOIUrl":"https://doi.org/10.1145/3501774.3501780","url":null,"abstract":"Within the past decade, the advent of cloud computing in terms of infrastructure, technology stacks, availability of services and tooling, along with the gradual improvement of its market environment, has driven many organizations to either consider or migrate many existing software systems to the cloud, either fully or partially. A common predicament in most cases, is the existence of a complex, monolithic application, potentially considered legacy at the time, that was not designed to be cloud-native and therefore requires a degree of redesign/reimplementation in order to benefit from cloud deployment. In such cases, the decomposition of the monolith to a set of loosely coupled, highly cohesive and self-contained microservices is a valid recommendation, provided that the organization is prepared to withstand the additional cost, in terms of human and financial resources, along with the unavoidable development overhead, which is inevitable during the early stages. However, the tendency of the tech world to embrace new trends and jump on hype trains for fear of obsoletion, has led to an excessive adoption of the microservices architecture (MSA), even in cases where such an architecture is not viable for the organization, or does not derive from any business requirements. This research focuses on establishing the position of a traditional monolith in the modern software architecture landscape and determine use cases that can still benefit from this paradigm, as well as use cases that could benefit from a partial or full transition to microservices architectures instead.","PeriodicalId":255059,"journal":{"name":"Proceedings of the 2021 European Symposium on Software Engineering","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129872230","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}
Andrew Flangas, J. Sattarvand, S. Dascalu, Fred Harris
This research entails using virtual reality to interpret video recordings in Unity from cameras on an unmanned machine used for mining excavations. The purpose of using a machine of this nature is to send it into hazardous mining environments rather than sending workers and having their lives jeopardized. This work is significant because it demonstrates how two separate fields, such as virtual reality and robotics, can be combined to complete useful tasks. It also illustrates how machines can be used to replace workers in hazardous conditions not only in the field of mining, but in other fields as well. The main contribution of the work presented in this paper is the creation of a panorama of live video feeds captured by several webcams, which can be seen using a VR headset. As also described in the paper the software developed for this engineering application has been created using appropriate software engineering techniques and tools. Results of merging live video feeds and testing camera placements are also presented and planned directions of future work are outlined.
{"title":"Merging Live Video Feeds for Remote Monitoring of a Mining Machine","authors":"Andrew Flangas, J. Sattarvand, S. Dascalu, Fred Harris","doi":"10.1145/3501774.3501776","DOIUrl":"https://doi.org/10.1145/3501774.3501776","url":null,"abstract":"This research entails using virtual reality to interpret video recordings in Unity from cameras on an unmanned machine used for mining excavations. The purpose of using a machine of this nature is to send it into hazardous mining environments rather than sending workers and having their lives jeopardized. This work is significant because it demonstrates how two separate fields, such as virtual reality and robotics, can be combined to complete useful tasks. It also illustrates how machines can be used to replace workers in hazardous conditions not only in the field of mining, but in other fields as well. The main contribution of the work presented in this paper is the creation of a panorama of live video feeds captured by several webcams, which can be seen using a VR headset. As also described in the paper the software developed for this engineering application has been created using appropriate software engineering techniques and tools. Results of merging live video feeds and testing camera placements are also presented and planned directions of future work are outlined.","PeriodicalId":255059,"journal":{"name":"Proceedings of the 2021 European Symposium on Software Engineering","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116965327","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 the development of society and the improvement of life quality, people's requirements for software experience are becoming more and more stringent. Performance optimization is becoming more and more important in software design. At present, most compilers support the default optimization options of o 1, o 2 and o 3, such as the GCC compiler. In order to meet the different degrees of optimization needs of users, it provides nearly a hundred optimization options for compiling time, target file length, and execution. At the same time, compiler optimization options will adjust the code execution order, and different optimization options will treat undefined behaviors differently. This work mainly tests the impact of undefined behavior (UB) on compiler optimization at different levels. The results show that UB has both positive and negative effects on the compiler. In addition, this paper reminds programmers to learn to improve their code specifications and correctly handle these undefined behaviors.
{"title":"The Impact of Undefined Behavior on Compiler Optimization","authors":"Zefan Shen","doi":"10.1145/3501774.3501781","DOIUrl":"https://doi.org/10.1145/3501774.3501781","url":null,"abstract":"With the development of society and the improvement of life quality, people's requirements for software experience are becoming more and more stringent. Performance optimization is becoming more and more important in software design. At present, most compilers support the default optimization options of o 1, o 2 and o 3, such as the GCC compiler. In order to meet the different degrees of optimization needs of users, it provides nearly a hundred optimization options for compiling time, target file length, and execution. At the same time, compiler optimization options will adjust the code execution order, and different optimization options will treat undefined behaviors differently. This work mainly tests the impact of undefined behavior (UB) on compiler optimization at different levels. The results show that UB has both positive and negative effects on the compiler. In addition, this paper reminds programmers to learn to improve their code specifications and correctly handle these undefined behaviors.","PeriodicalId":255059,"journal":{"name":"Proceedings of the 2021 European Symposium on Software Engineering","volume":"128 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132686227","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}
A. Tahat, Rozana Awwad, Nadia Baydoun, Shurooq Al-Nabih, T. Edwan
WiFi based indoor localization systems relying on received signal strength indicator (RSSI) have attained large acceptance over the past few years as mobile devices with WiFi capability are prevalent in everyday routines and practices, this is in view of the demand for low-cost indoor positioning systems (IPSs). Extant RSSI reliant WiFi based IPSs utilize raw RSSI of signals received from access points (APs) to evaluate device locations. However, the particular raw RSSI of signals may readily fluctuate and may be susceptible to interference as a consequence to multipath propagation channels, other wireless local area networks, device diverseness, and noise. To overcome these prevailing issues, we investigate performance enhancement of WiFi fingerprinting-based IPSs for increased accuracy and robustness in positioning over dual-band WiFi (such as IEEE 802.11n) that employs both of the 2.4GHz and 5GHz frequency spectrum bands as a conceivable substitute. To that end, based on empirical concurrent measurements, we conduct a comparative performance analysis of a collection of machine learning (ML) classification algorithms to evaluate their classification capacities in determining the location of a WiFi receiver device in a single and a dual frequency band operation setting. Numerical results demonstrated that in our IPS, the location could be effectively predicted by means of a subset of the collection of considered ML classification algorithms when using the raw RSSI of both of the 2.4GHz and 5GHz frequency bands. Computed evaluation metrics to characterize performance of the IPS served to identify an optimum ML algorithm based on attained results to accurately localize the device, and the existence of dual frequency information renders the positioning process to be more robust.
{"title":"An Empirical Evaluation of Machine Learning Algorithms for Indoor Localization using Dual-Band WiFi","authors":"A. Tahat, Rozana Awwad, Nadia Baydoun, Shurooq Al-Nabih, T. Edwan","doi":"10.1145/3501774.3501790","DOIUrl":"https://doi.org/10.1145/3501774.3501790","url":null,"abstract":"WiFi based indoor localization systems relying on received signal strength indicator (RSSI) have attained large acceptance over the past few years as mobile devices with WiFi capability are prevalent in everyday routines and practices, this is in view of the demand for low-cost indoor positioning systems (IPSs). Extant RSSI reliant WiFi based IPSs utilize raw RSSI of signals received from access points (APs) to evaluate device locations. However, the particular raw RSSI of signals may readily fluctuate and may be susceptible to interference as a consequence to multipath propagation channels, other wireless local area networks, device diverseness, and noise. To overcome these prevailing issues, we investigate performance enhancement of WiFi fingerprinting-based IPSs for increased accuracy and robustness in positioning over dual-band WiFi (such as IEEE 802.11n) that employs both of the 2.4GHz and 5GHz frequency spectrum bands as a conceivable substitute. To that end, based on empirical concurrent measurements, we conduct a comparative performance analysis of a collection of machine learning (ML) classification algorithms to evaluate their classification capacities in determining the location of a WiFi receiver device in a single and a dual frequency band operation setting. Numerical results demonstrated that in our IPS, the location could be effectively predicted by means of a subset of the collection of considered ML classification algorithms when using the raw RSSI of both of the 2.4GHz and 5GHz frequency bands. Computed evaluation metrics to characterize performance of the IPS served to identify an optimum ML algorithm based on attained results to accurately localize the device, and the existence of dual frequency information renders the positioning process to be more robust.","PeriodicalId":255059,"journal":{"name":"Proceedings of the 2021 European Symposium on Software Engineering","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132102789","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}
N. Petrellis, Stavros Zogas, P. Christakos, Panagiotis G. Mousouliotis, G. Keramidas, N. Voros, Christos D. Antonopoulos
Shape tracking is based on landmark detection and alignment. Open-source code and pre-trained models are available for an implementation that is based on an ensemble of regression trees. The C++ Deformable Shape Tracking (DEST) implementation of face alignment that is using Eigen template library for algebraic operations is employed in this work. The overhead of the C++ Eigen library calls is measured and selected computational intensive operations are ported from Eigen implementation to custom C code achieving a remarkable acceleration in the shape tracking application. An important achievement of this work is the fact that the restructured code can be directly implemented with reconfigurable hardware for further speed improvement. Driver drowsiness and distraction detection applications are exploiting shape tracking by measuring landmark distances in order to detect eye blinking, yawning, etc. Fast video processing and accuracy is mandatory in these safety critical applications. The modified software implementation of the original DEST face alignment method presented in this paper, is almost 250 times faster due to the custom implementation of computational intensive vector/matrix operations and rotations. Eigen library is still used in non-time critical parts of the code for compact description and higher readability. Flattening of nested routines and inline implementation is also used to eliminate excessive argument copies and data type checking and conversions.
{"title":"Software Acceleration of the Deformable Shape Tracking Application: How to eliminate the Eigen Library Overhead","authors":"N. Petrellis, Stavros Zogas, P. Christakos, Panagiotis G. Mousouliotis, G. Keramidas, N. Voros, Christos D. Antonopoulos","doi":"10.1145/3501774.3501782","DOIUrl":"https://doi.org/10.1145/3501774.3501782","url":null,"abstract":"Shape tracking is based on landmark detection and alignment. Open-source code and pre-trained models are available for an implementation that is based on an ensemble of regression trees. The C++ Deformable Shape Tracking (DEST) implementation of face alignment that is using Eigen template library for algebraic operations is employed in this work. The overhead of the C++ Eigen library calls is measured and selected computational intensive operations are ported from Eigen implementation to custom C code achieving a remarkable acceleration in the shape tracking application. An important achievement of this work is the fact that the restructured code can be directly implemented with reconfigurable hardware for further speed improvement. Driver drowsiness and distraction detection applications are exploiting shape tracking by measuring landmark distances in order to detect eye blinking, yawning, etc. Fast video processing and accuracy is mandatory in these safety critical applications. The modified software implementation of the original DEST face alignment method presented in this paper, is almost 250 times faster due to the custom implementation of computational intensive vector/matrix operations and rotations. Eigen library is still used in non-time critical parts of the code for compact description and higher readability. Flattening of nested routines and inline implementation is also used to eliminate excessive argument copies and data type checking and conversions.","PeriodicalId":255059,"journal":{"name":"Proceedings of the 2021 European Symposium on Software Engineering","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127474528","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：