2021 |
Edraki, Amin; Chan, Wai-Yip; Jensen, Jesper; Fogerty, Daniel Speech Intelligibility Prediction Using Spectro-Temporal Modulation Analysis Journal Article IEEE/ACM Transactions on Audio, Speech, and Language Processing, 29 , pp. 210-225, 2021. @article{9269417, title = {Speech Intelligibility Prediction Using Spectro-Temporal Modulation Analysis}, author = {Amin Edraki and Wai-Yip Chan and Jesper Jensen and Daniel Fogerty}, doi = {10.1109/TASLP.2020.3039929}, year = {2021}, date = {2021-11-24}, journal = {IEEE/ACM Transactions on Audio, Speech, and Language Processing}, volume = {29}, pages = {210-225}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Lei, Jiarui; Fan, Dixia; Angera, Andrea; Liu, Yuming; Nepf, Heidi Drag force and reconfiguration of cultivated Saccharina latissima in current Journal Article Aquacultural Engineering, 94 , pp. 102169, 2021, ISSN: 0144-8609. Abstract | Links | BibTeX | Tags: Aquaculture, Drag force, Flow-vegetation interaction, Kelp, Reconfiguration @article{LEI2021102169, title = {Drag force and reconfiguration of cultivated Saccharina latissima in current}, author = {Jiarui Lei and Dixia Fan and Andrea Angera and Yuming Liu and Heidi Nepf}, url = {https://www.sciencedirect.com/science/article/pii/S014486092100025X}, doi = {https://doi.org/10.1016/j.aquaeng.2021.102169}, issn = {0144-8609}, year = {2021}, date = {2021-08-01}, journal = {Aquacultural Engineering}, volume = {94}, pages = {102169}, abstract = {The design of aquaculture systems requires an understanding of the drag forces on cultivated kelp. This study measured the drag on line segments of cultivated Saccharina latissima in a towing tank. The drag on segments of farm line with full kelp bundles and with stipes alone (fronds removed) was measured at tow speeds of 0.10 to 0.50 m/s. The drag on individual fronds cut from the line was also measured. Video images were collected to evaluate the plant reconfiguration. Both kelp blades and stipes contributed to the total drag force on the line bundle. Within the velocity range of our experiments, the kelp blades were essentially horizontal. However, the pronation of kelp stipes increased as flow velocity increased. The reconfiguration of kelp stipes was observed to decrease the vertical extent of the kelp bundle. Due to this reconfiguration, the measured force, F, increased with velocity, U, at a rate slower than quadratic, and was consistent with scaling laws derived for reconfiguration. Specifically, F∼Uα with α=1.35±0.17.}, keywords = {Aquaculture, Drag force, Flow-vegetation interaction, Kelp, Reconfiguration}, pubstate = {published}, tppubtype = {article} } The design of aquaculture systems requires an understanding of the drag forces on cultivated kelp. This study measured the drag on line segments of cultivated Saccharina latissima in a towing tank. The drag on segments of farm line with full kelp bundles and with stipes alone (fronds removed) was measured at tow speeds of 0.10 to 0.50 m/s. The drag on individual fronds cut from the line was also measured. Video images were collected to evaluate the plant reconfiguration. Both kelp blades and stipes contributed to the total drag force on the line bundle. Within the velocity range of our experiments, the kelp blades were essentially horizontal. However, the pronation of kelp stipes increased as flow velocity increased. The reconfiguration of kelp stipes was observed to decrease the vertical extent of the kelp bundle. Due to this reconfiguration, the measured force, F, increased with velocity, U, at a rate slower than quadratic, and was consistent with scaling laws derived for reconfiguration. Specifically, F∼Uα with α=1.35±0.17. |
Mokogwu, Chiedu N; Hashtrudi-Zaad, Keyvan A hybrid position–rate teleoperation system Journal Article Robotics and Autonomous Systems, 141 , pp. 103781, 2021, ISSN: 0921-8890. Abstract | Links | BibTeX | Tags: Bilateral teleoperation, Position mode, Range sensor, Rate mode, Stability, Workspace expansion @article{MOKOGWU2021103781, title = {A hybrid position–rate teleoperation system}, author = {Chiedu N Mokogwu and Keyvan Hashtrudi-Zaad}, url = {https://www.sciencedirect.com/science/article/pii/S092188902100066X}, doi = {https://doi.org/10.1016/j.robot.2021.103781}, issn = {0921-8890}, year = {2021}, date = {2021-07-01}, journal = {Robotics and Autonomous Systems}, volume = {141}, pages = {103781}, abstract = {Position resolution is a major problem in teleoperation applications with significant disparity between master and slave workspace. While rate mode control is suitable for slave free motion operations, it poses significant stability and performance challenges for task manipulation. In this paper, we propose a hybrid control scheme that offers seamless transition between position and rate control modes based on the environment location information collected from a range sensor. The system incorporates the strengths of position and rate control modes while masking their shortcomings. Experiments to determine the viability of this method are carried out on a single degree-of-freedom teleoperation test-bed.}, keywords = {Bilateral teleoperation, Position mode, Range sensor, Rate mode, Stability, Workspace expansion}, pubstate = {published}, tppubtype = {article} } Position resolution is a major problem in teleoperation applications with significant disparity between master and slave workspace. While rate mode control is suitable for slave free motion operations, it poses significant stability and performance challenges for task manipulation. In this paper, we propose a hybrid control scheme that offers seamless transition between position and rate control modes based on the environment location information collected from a range sensor. The system incorporates the strengths of position and rate control modes while masking their shortcomings. Experiments to determine the viability of this method are carried out on a single degree-of-freedom teleoperation test-bed. |
Sato, Thiago H; dos Santos, Sérgio Barros R; de Oliveira, André M; Cappabianco, Fábio A M; Givigi, Sidney N Experimental Validation of a Steering Control System using an Adaptive Fuzzy Controller and Computer Vision Inproceedings 2021 IEEE International Systems Conference (SysCon), pp. 1-8, 2021. @inproceedings{9447097, title = {Experimental Validation of a Steering Control System using an Adaptive Fuzzy Controller and Computer Vision}, author = {Thiago H Sato and Sérgio Barros R dos Santos and André M de Oliveira and Fábio A M Cappabianco and Sidney N Givigi}, doi = {10.1109/SysCon48628.2021.9447097}, year = {2021}, date = {2021-06-09}, booktitle = {2021 IEEE International Systems Conference (SysCon)}, pages = {1-8}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } |
Magee, Riley; Givigi, Sidney A System For P300 Detection Applied To Vehicle Navigation Inproceedings 2021 IEEE International Systems Conference (SysCon), pp. 1-8, 2021. @inproceedings{9447129, title = {A System For P300 Detection Applied To Vehicle Navigation}, author = {Riley Magee and Sidney Givigi}, doi = {10.1109/SysCon48628.2021.9447129}, year = {2021}, date = {2021-06-09}, booktitle = {2021 IEEE International Systems Conference (SysCon)}, pages = {1-8}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } |
Heaton, Jacqueline; Givigi, Sidney Systems Design for EEG Signal Classification of Sensorimotor Activity Using Machine Learning Inproceedings 2021 IEEE International Systems Conference (SysCon), pp. 1-7, 2021. @inproceedings{9447106, title = {Systems Design for EEG Signal Classification of Sensorimotor Activity Using Machine Learning}, author = {Jacqueline Heaton and Sidney Givigi}, doi = {10.1109/SysCon48628.2021.9447106}, year = {2021}, date = {2021-06-09}, booktitle = {2021 IEEE International Systems Conference (SysCon)}, pages = {1-7}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } |
Campos, José Creissac; Graham, Nicholas T C; Spano, Lucio Davide; den Bergh, Jan Van The Management of Risks and Benefits When Engineering Interactive Digital Systems Inproceedings Companion of the 2021 ACM SIGCHI Symposium on Engineering Interactive Computing Systems, pp. 81–83, Association for Computing Machinery, Virtual Event, Netherlands, 2021, ISBN: 9781450384490. Abstract | Links | BibTeX | Tags: engineering interactive systems, risk management, safety-critical @inproceedings{10.1145/3459926.3464760, title = {The Management of Risks and Benefits When Engineering Interactive Digital Systems}, author = {José Creissac Campos and Nicholas T C Graham and Lucio Davide Spano and Jan Van den Bergh}, url = {https://doi.org/10.1145/3459926.3464760}, doi = {10.1145/3459926.3464760}, isbn = {9781450384490}, year = {2021}, date = {2021-06-08}, booktitle = {Companion of the 2021 ACM SIGCHI Symposium on Engineering Interactive Computing Systems}, pages = {81–83}, publisher = {Association for Computing Machinery}, address = {Virtual Event, Netherlands}, series = {EICS '21}, abstract = {Traditionally, most UX designers, computer scientists and software engineers have not had to consider risks to the public from using their systems. However, the current evolution of digital systems in terms of the increasing number of users, their growing complexity and the pervasiveness of Artificial Intelligence techniques allow common HCI designers and engineers to build systems that create risks for the individual, groups of people, or event to the entire society. In this workshop, we aim at collecting the views and the current practice in the management of the risks and benefits in the engineering of interactive digital systems. Such a view will draw the way for new research, methods, and tools to incorporate the risk analysis into the current engineering and design practices. The workshop is proposed on behalf of the IFIP Working Groups 2.7/13.4 on User Interface Engineering.}, keywords = {engineering interactive systems, risk management, safety-critical}, pubstate = {published}, tppubtype = {inproceedings} } Traditionally, most UX designers, computer scientists and software engineers have not had to consider risks to the public from using their systems. However, the current evolution of digital systems in terms of the increasing number of users, their growing complexity and the pervasiveness of Artificial Intelligence techniques allow common HCI designers and engineers to build systems that create risks for the individual, groups of people, or event to the entire society. In this workshop, we aim at collecting the views and the current practice in the management of the risks and benefits in the engineering of interactive digital systems. Such a view will draw the way for new research, methods, and tools to incorporate the risk analysis into the current engineering and design practices. The workshop is proposed on behalf of the IFIP Working Groups 2.7/13.4 on User Interface Engineering. |
Davoodnia, Vandad; Ghorbani, Saeed; Etemad, Ali Estimating pose from pressure data for smart beds with deep image-based pose estimators Journal Article Applied Intelligence, 2021, ISBN: Davoodnia2021. Abstract | Links | BibTeX | Tags: Machine Learning, Smart Beds @article{Davoodnia2021, title = {Estimating pose from pressure data for smart beds with deep image-based pose estimators}, author = {Vandad Davoodnia and Saeed Ghorbani and Ali Etemad }, url = {https://link.springer.com/article/10.1007/s10489-021-02418-y}, doi = {https://doi.org/10.1007/s10489-021-02418-y}, isbn = {Davoodnia2021}, year = {2021}, date = {2021-06-04}, journal = {Applied Intelligence}, abstract = {In-bed pose estimation has shown value in fields such as hospital patient monitoring, sleep studies, and smart homes. In this paper, we explore different strategies for detecting body pose from highly ambiguous pressure data, with the aid of pre-existing pose estimators. We examine the performance of pre-trained pose estimators by using them either directly or by re-training them on two pressure datasets. We also explore other strategies utilizing a learnable pre-processing domain adaptation step, which transforms the vague pressure maps to a representation closer to the expected input space of common purpose pose estimation modules. Accordingly, we used a fully convolutional network with multiple scales to provide the pose-specific characteristics of the pressure maps to the pre-trained pose estimation module. Our complete analysis of different approaches shows that the combination of learnable pre-processing module along with re-training pre-existing image-based pose estimators on the pressure data is able to overcome issues such as highly vague pressure points to achieve very high pose estimation accuracy.}, keywords = {Machine Learning, Smart Beds}, pubstate = {published}, tppubtype = {article} } In-bed pose estimation has shown value in fields such as hospital patient monitoring, sleep studies, and smart homes. In this paper, we explore different strategies for detecting body pose from highly ambiguous pressure data, with the aid of pre-existing pose estimators. We examine the performance of pre-trained pose estimators by using them either directly or by re-training them on two pressure datasets. We also explore other strategies utilizing a learnable pre-processing domain adaptation step, which transforms the vague pressure maps to a representation closer to the expected input space of common purpose pose estimation modules. Accordingly, we used a fully convolutional network with multiple scales to provide the pose-specific characteristics of the pressure maps to the pre-trained pose estimation module. Our complete analysis of different approaches shows that the combination of learnable pre-processing module along with re-training pre-existing image-based pose estimators on the pressure data is able to overcome issues such as highly vague pressure points to achieve very high pose estimation accuracy. |
Lin, Ke (林柯 ); Wang, Jiasong (王嘉松 ); Fan, Dixia (范迪夏 ); Triantafyllou, Michael S Flow-induced cross-flow vibrations of long flexible cylinder with an upstream wake interference Journal Article Physics of Fluids, 33 (6), pp. 065104, 2021. @article{doi:10.1063/5.0053826, title = {Flow-induced cross-flow vibrations of long flexible cylinder with an upstream wake interference}, author = {Ke (林柯 ) Lin and Jiasong (王嘉松 ) Wang and Dixia (范迪夏 ) Fan and Michael S Triantafyllou}, url = {https://doi.org/10.1063/5.0053826}, doi = {10.1063/5.0053826}, year = {2021}, date = {2021-06-02}, journal = {Physics of Fluids}, volume = {33}, number = {6}, pages = {065104}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Dai, Yinpei; Li, Hangyu; Li, Yongbin; Sun, Jian; Huang, Fei; Si, Luo; Zhu, Xiaodan Preview, Attend and Review: Schema-Aware Curriculum Learning for Multi-Domain Dialog State Tracking Miscellaneous 2021. BibTeX | Tags: @misc{dai2021preview, title = {Preview, Attend and Review: Schema-Aware Curriculum Learning for Multi-Domain Dialog State Tracking}, author = {Yinpei Dai and Hangyu Li and Yongbin Li and Jian Sun and Fei Huang and Luo Si and Xiaodan Zhu}, year = {2021}, date = {2021-06-01}, keywords = {}, pubstate = {published}, tppubtype = {misc} } |
Shepertycky, Michael; Burton, Sarah; Dickson, Andrew; Liu, Yan-Fei; Li, Qingguo Removing energy with an exoskeleton reduces the metabolic cost of walking Journal Article Science, 372 (6545), pp. 957–960, 2021, ISSN: 0036-8075. Abstract | Links | BibTeX | Tags: @article{Shepertycky957, title = {Removing energy with an exoskeleton reduces the metabolic cost of walking}, author = {Michael Shepertycky and Sarah Burton and Andrew Dickson and Yan-Fei Liu and Qingguo Li}, url = {https://science.sciencemag.org/content/372/6545/957}, doi = {10.1126/science.aba9947}, issn = {0036-8075}, year = {2021}, date = {2021-05-28}, journal = {Science}, volume = {372}, number = {6545}, pages = {957--960}, publisher = {American Association for the Advancement of Science}, abstract = {Many devices have been developed to harvest energy from walking or running, but their use often comes at cost to the wearer in the form of increased metabolic demand. Shepertycky et al. designed a device that can harvest mechanical energy from a natural walking gait and convert it to useable electrical energy while also reducing the metabolic energy consumption of the user (see the Perspective by Riemer et al.). The key to achieving textquotedblleftsomething from nothingtextquotedblright comes from designing the device to use muscle-centric control of the knee exoskeleton resistance to reduce active muscle force during the late part of the leg swing cycle.Science, aba9947, this issue p. 957; see also abh4007, p. 909Evolutionary pressures have led humans to walk in a highly efficient manner that conserves energy, making it difficult for exoskeletons to reduce the metabolic cost of walking. Despite the challenge, some exoskeletons have managed to lessen the metabolic expenditure of walking, either by adding or storing and returning energy. We show that the use of an exoskeleton that strategically removes kinetic energy during the swing period of the gait cycle reduces the metabolic cost of walking by 2.5 textpm 0.8% for healthy male users while converting the removed energy into 0.25 textpm 0.02 watts of electrical power. By comparing two loading profiles, we demonstrate that the timing and magnitude of energy removal are vital for successful metabolic cost reduction.}, keywords = {}, pubstate = {published}, tppubtype = {article} } Many devices have been developed to harvest energy from walking or running, but their use often comes at cost to the wearer in the form of increased metabolic demand. Shepertycky et al. designed a device that can harvest mechanical energy from a natural walking gait and convert it to useable electrical energy while also reducing the metabolic energy consumption of the user (see the Perspective by Riemer et al.). The key to achieving textquotedblleftsomething from nothingtextquotedblright comes from designing the device to use muscle-centric control of the knee exoskeleton resistance to reduce active muscle force during the late part of the leg swing cycle.Science, aba9947, this issue p. 957; see also abh4007, p. 909Evolutionary pressures have led humans to walk in a highly efficient manner that conserves energy, making it difficult for exoskeletons to reduce the metabolic cost of walking. Despite the challenge, some exoskeletons have managed to lessen the metabolic expenditure of walking, either by adding or storing and returning energy. We show that the use of an exoskeleton that strategically removes kinetic energy during the swing period of the gait cycle reduces the metabolic cost of walking by 2.5 textpm 0.8% for healthy male users while converting the removed energy into 0.25 textpm 0.02 watts of electrical power. By comparing two loading profiles, we demonstrate that the timing and magnitude of energy removal are vital for successful metabolic cost reduction. |
CONTEXT BASED PREDICTION OF A COMMUNICATION TIME Journal Article (20210150381), 2021. @article{patent:20210150381, title = {CONTEXT BASED PREDICTION OF A COMMUNICATION TIME}, url = {https://www.freepatentsonline.com/y2021/0150381.html}, year = {2021}, date = {2021-05-20}, number = {20210150381}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Gu, Jia-Chen; Liu, Hui; Ling, Zhen-Hua; Liu, Quan; Chen, Zhigang; Zhu, Xiaodan Partner Matters! An Empirical Study on Fusing Personas for Personalized Response Selection in Retrieval-Based Chatbots Miscellaneous 2021. BibTeX | Tags: @misc{gu2021partner, title = {Partner Matters! An Empirical Study on Fusing Personas for Personalized Response Selection in Retrieval-Based Chatbots}, author = {Jia-Chen Gu and Hui Liu and Zhen-Hua Ling and Quan Liu and Zhigang Chen and Xiaodan Zhu}, year = {2021}, date = {2021-05-19}, keywords = {}, pubstate = {published}, tppubtype = {misc} } |
Zhang, Chuan; Zhu, Liehuang; Xu, Chang; Ni, Jianbing; Huang, Cheng; Shen, Xuemin Sherman Location Privacy-Preserving Task Recommendation with Geometric Range Query in Mobile Crowdsensing Journal Article IEEE Transactions on Mobile Computing, pp. 1-1, 2021. @article{9432741, title = {Location Privacy-Preserving Task Recommendation with Geometric Range Query in Mobile Crowdsensing}, author = {Chuan Zhang and Liehuang Zhu and Chang Xu and Jianbing Ni and Cheng Huang and Xuemin Sherman Shen}, doi = {10.1109/TMC.2021.3080714}, year = {2021}, date = {2021-05-17}, journal = {IEEE Transactions on Mobile Computing}, pages = {1-1}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Li, Tianda; Gu, Jia-Chen; Liu, Hui; Liu, Quan; Ling, Zhen-Hua; Su, Zhiming; Zhu, Xiaodan Have You Made a Decision? Where? A Pilot Study on Interpretability of Polarity Analysis Based on Advising Problem Inproceedings ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 6928-6932, 2021. @inproceedings{9413654, title = {Have You Made a Decision? Where? A Pilot Study on Interpretability of Polarity Analysis Based on Advising Problem}, author = {Tianda Li and Jia-Chen Gu and Hui Liu and Quan Liu and Zhen-Hua Ling and Zhiming Su and Xiaodan Zhu}, doi = {10.1109/ICASSP39728.2021.9413654}, year = {2021}, date = {2021-05-13}, booktitle = {ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}, pages = {6928-6932}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } |
Sepas-Moghaddam, Alireza; Pereira, Fernando; Correia, Paulo Lobato; Etemad, Ali Multi-Perspective LSTM for Joint Visual Representation Learning Miscellaneous 2021. BibTeX | Tags: @misc{sepasmoghaddam2021multiperspective, title = {Multi-Perspective LSTM for Joint Visual Representation Learning}, author = {Alireza Sepas-Moghaddam and Fernando Pereira and Paulo Lobato Correia and Ali Etemad}, year = {2021}, date = {2021-05-06}, keywords = {}, pubstate = {published}, tppubtype = {misc} } |
Yager, Jacob; Hoult, Neil A; Bentz, Evan Evaluating the behaviour of functionally graded reinforced concrete without transverse reinforcement using distributed sensing Journal Article Construction and Building Materials, 295 , pp. 123612, 2021, ISSN: 0950-0618. Abstract | Links | BibTeX | Tags: Debonding, Digital image correlation, Distributed fibre optic strain sensing, Fibre reinforced concrete, Fresh on fresh concrete casting, Functionally graded concrete, Low cement concrete, Three-point bending tests @article{YAGER2021123612, title = {Evaluating the behaviour of functionally graded reinforced concrete without transverse reinforcement using distributed sensing}, author = {Jacob Yager and Neil A Hoult and Evan Bentz}, url = {https://www.sciencedirect.com/science/article/pii/S0950061821013726}, doi = {https://doi.org/10.1016/j.conbuildmat.2021.123612}, issn = {0950-0618}, year = {2021}, date = {2021-05-01}, journal = {Construction and Building Materials}, volume = {295}, pages = {123612}, abstract = {Functionally graded concrete (FGC) can lower cement content and CO2 emissions. However, the behaviour of FGC is not fully understood. To examine behaviour at concrete interfaces as well as global behaviour, horizontally and vertically layered FGC (HLFGC, VLFGC) and control beams, all without transverse reinforcement, were tested in three-point bending. Distributed sensing measured reinforcement strains, cracking behaviour, and load–deflection behaviour. Using fibre reinforced concrete increased the load carrying capacity of VLFGC and decreased the number of cracks and the cracking load for HLFGC. Cracking at interfaces produced lower cracking strengths in VLFGC beams and brittle failures in some HLFGC beams.}, keywords = {Debonding, Digital image correlation, Distributed fibre optic strain sensing, Fibre reinforced concrete, Fresh on fresh concrete casting, Functionally graded concrete, Low cement concrete, Three-point bending tests}, pubstate = {published}, tppubtype = {article} } Functionally graded concrete (FGC) can lower cement content and CO2 emissions. However, the behaviour of FGC is not fully understood. To examine behaviour at concrete interfaces as well as global behaviour, horizontally and vertically layered FGC (HLFGC, VLFGC) and control beams, all without transverse reinforcement, were tested in three-point bending. Distributed sensing measured reinforcement strains, cracking behaviour, and load–deflection behaviour. Using fibre reinforced concrete increased the load carrying capacity of VLFGC and decreased the number of cracks and the cracking load for HLFGC. Cracking at interfaces produced lower cracking strengths in VLFGC beams and brittle failures in some HLFGC beams. |
Zand, Mohsen; Etemad, Ali; Greenspan, Michael Oriented Bounding Boxes for Small and Freely Rotated Objects Miscellaneous 2021. @misc{zand2021oriented, title = {Oriented Bounding Boxes for Small and Freely Rotated Objects}, author = {Mohsen Zand and Ali Etemad and Michael Greenspan}, url = {https://arxiv.org/pdf/2104.11854.pdf}, year = {2021}, date = {2021-04-21}, keywords = {}, pubstate = {published}, tppubtype = {misc} } |
Zand, Mohsen; Etemad, Ali; Greenspan, Michael Flow-based Autoregressive Structured Prediction of Human Motion Miscellaneous 2021. @misc{zand2021flowbased, title = {Flow-based Autoregressive Structured Prediction of Human Motion}, author = {Mohsen Zand and Ali Etemad and Michael Greenspan}, url = {https://arxiv.org/pdf/2104.04391.pdf}, year = {2021}, date = {2021-04-09}, keywords = {}, pubstate = {published}, tppubtype = {misc} } |
Shah, Hurmat Ali; Zhao, Lian; Kim, Il-Min IEEE Transactions on Vehicular Technology, pp. 1-1, 2021. @article{9399261, title = {Joint Network Control and Resource Allocation for Space-Terrestrial Integrated Network through Hierarchal Deep Actor-Critic Reinforcement Learning}, author = {Hurmat Ali Shah and Lian Zhao and Il-Min Kim}, doi = {10.1109/TVT.2021.3071983}, year = {2021}, date = {2021-04-08}, journal = {IEEE Transactions on Vehicular Technology}, pages = {1-1}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Wu, Yangzheng; Zand, Mohsen; Etemad, Ali; Greenspan, Michael Vote from the Center: 6 DoF Pose Estimation in RGB-D Images by Radial Keypoint Voting Miscellaneous 2021. @misc{wu2021vote, title = {Vote from the Center: 6 DoF Pose Estimation in RGB-D Images by Radial Keypoint Voting}, author = {Yangzheng Wu and Mohsen Zand and Ali Etemad and Michael Greenspan}, url = {https://arxiv.org/pdf/2104.02527.pdf}, year = {2021}, date = {2021-04-06}, keywords = {}, pubstate = {published}, tppubtype = {misc} } |
Uppal, Hardik; Sepas-Moghaddam, Alireza; Greenspan, Michael; Etemad, Ali Teacher-Student Adversarial Depth Hallucination to Improve Face Recognition Miscellaneous 2021. @misc{uppal2021teacherstudent, title = {Teacher-Student Adversarial Depth Hallucination to Improve Face Recognition}, author = {Hardik Uppal and Alireza Sepas-Moghaddam and Michael Greenspan and Ali Etemad}, url = {https://arxiv.org/pdf/2104.02424.pdf}, year = {2021}, date = {2021-04-06}, keywords = {}, pubstate = {published}, tppubtype = {misc} } |
Yun, S; Kang, J -M; Ha, J; Lee, S; Ryu, D -W; Kwon, J; Kim, I -M Deep Learning-Based Ground Vibration Monitoring: Reinforcement Learning and RNN-CNN Approach Journal Article IEEE Geoscience and Remote Sensing Letters, pp. 1-5, 2021. @article{9389744, title = {Deep Learning-Based Ground Vibration Monitoring: Reinforcement Learning and RNN-CNN Approach}, author = {S Yun and J -M Kang and J Ha and S Lee and D -W Ryu and J Kwon and I -M Kim}, doi = {10.1109/LGRS.2021.3067974}, year = {2021}, date = {2021-03-30}, journal = {IEEE Geoscience and Remote Sensing Letters}, pages = {1-5}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Nair, M; Givigi, S Communications design in multi-robotic systems Inproceedings 2020 International Conference on Communications, Signal Processing, and their Applications (ICCSPA), pp. 1-6, 2021. @inproceedings{9385740, title = {Communications design in multi-robotic systems}, author = {M Nair and S Givigi}, doi = {10.1109/ICCSPA49915.2021.9385740}, year = {2021}, date = {2021-03-21}, booktitle = {2020 International Conference on Communications, Signal Processing, and their Applications (ICCSPA)}, pages = {1-6}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } |
Jeong, J; Kim, I -M; Hong, D Deep Reinforcement Learning-based Task Offloading Decision in the Time Varying Channel Inproceedings 2021 International Conference on Electronics, Information, and Communication (ICEIC), pp. 1-4, 2021. Abstract | Links | BibTeX | Tags: Heuristic algorithms;Computational modeling;Simulation;Reinforcement learning;Delays;Coherence time;Task analysis;Computation offloading;multi-access edge computing;reinforcement learning;DQN @inproceedings{9369737, title = {Deep Reinforcement Learning-based Task Offloading Decision in the Time Varying Channel}, author = {J Jeong and I -M Kim and D Hong}, doi = {10.1109/ICEIC51217.2021.9369737}, year = {2021}, date = {2021-03-10}, booktitle = {2021 International Conference on Electronics, Information, and Communication (ICEIC)}, pages = {1-4}, abstract = {This paper proposes a dynamic task offloading decision control scheme to minimize the total delay to execute computation task taking into account the time-varying channel. Specifically, we consider the practical task offloading process, where executing computation task is carried out over multiple channel coherence times. In order to make an accurate decision on the task offloading process performed over multiple channel coherence times, we utilize the model-free reinforcement learning, since environment dynamics of the system, channel transition probabilities, is challenging to estimate. We formulate a problem of minimizing the total delay of executing computation task based on a Markov decision process (MDP). In order to solve the MDP problem, we develop a model-free reinforcement learning algorithm. Simulation results show that our proposed scheme outperforms the conventional scheme.}, keywords = {Heuristic algorithms;Computational modeling;Simulation;Reinforcement learning;Delays;Coherence time;Task analysis;Computation offloading;multi-access edge computing;reinforcement learning;DQN}, pubstate = {published}, tppubtype = {inproceedings} } This paper proposes a dynamic task offloading decision control scheme to minimize the total delay to execute computation task taking into account the time-varying channel. Specifically, we consider the practical task offloading process, where executing computation task is carried out over multiple channel coherence times. In order to make an accurate decision on the task offloading process performed over multiple channel coherence times, we utilize the model-free reinforcement learning, since environment dynamics of the system, channel transition probabilities, is challenging to estimate. We formulate a problem of minimizing the total delay of executing computation task based on a Markov decision process (MDP). In order to solve the MDP problem, we develop a model-free reinforcement learning algorithm. Simulation results show that our proposed scheme outperforms the conventional scheme. |
Hui, Binyuan; Shi, Xiang; Geng, Ruiying; Li, Binhua; Li, Yongbin; Sun, Jian; Zhu, Xiaodan Improving Text-to-SQL with Schema Dependency Learning Online 2021. BibTeX | Tags: @online{hui2021improving, title = {Improving Text-to-SQL with Schema Dependency Learning}, author = {Binyuan Hui and Xiang Shi and Ruiying Geng and Binhua Li and Yongbin Li and Jian Sun and Xiaodan Zhu}, year = {2021}, date = {2021-03-07}, keywords = {}, pubstate = {published}, tppubtype = {online} } |
Yan, Ruobing; Take, Andy W; Hoult, Neil A; Meehan, Jonathan; Levesque, Christiane Evaluation of Shape Array sensors to quantify the spatial distribution and seasonal rate of track settlement Journal Article Transportation Geotechnics, 27 , pp. 100487, 2021, ISSN: 2214-3912. Abstract | Links | BibTeX | Tags: Field Test, LIDAR, Rail settlement monitoring, Shape Array (SAA), Track geometry irregularities @article{YAN2021100487, title = {Evaluation of Shape Array sensors to quantify the spatial distribution and seasonal rate of track settlement}, author = {Ruobing Yan and Andy W Take and Neil A Hoult and Jonathan Meehan and Christiane Levesque}, url = {http://www.sciencedirect.com/science/article/pii/S2214391220303755}, doi = {https://doi.org/10.1016/j.trgeo.2020.100487}, issn = {2214-3912}, year = {2021}, date = {2021-03-01}, journal = {Transportation Geotechnics}, volume = {27}, pages = {100487}, abstract = {Rail track geometry irregularities can lead to ride discomfort for passengers and redistribution of wheel loads potentially causing derailments. Current techniques for monitoring track settlements involve the use of discrete sensors or vehicle mounted sensors, which make it difficult to capture either spatial or temporal variations in settlement. Shape Array sensors (SAA) can potentially be used to capture temporal variations in distributed track settlement profiles to monitor and investigate potential track geometry irregularities and inform track maintenance programs. In this study, a site with known ground deformation issues (i.e. soft spots) was monitored with both an SAA and intermittent LIDAR scans. The objectives of the research were to investigate the accuracy of the SAA for measuring track settlements based on a comparison with LIDAR data, use those measurements to assess track irregularities at the site, evaluate temporal and seasonal changes in rail deformation, and to gain insight into the underlying causes of the ground deformation issues. The SAA was found to provide comparable settlement measurements to those from the LIDAR with the added advantage that the data could be used to assess settlement rates. At this particular site, long-term rail settlements were found to be a function of seasonal/climatic conditions with slower settlement rates in the winter and higher rates in the summer. In addition, the measurements indicated that the deformation issues are potentially caused by an asymmetric bearing failure.}, keywords = {Field Test, LIDAR, Rail settlement monitoring, Shape Array (SAA), Track geometry irregularities}, pubstate = {published}, tppubtype = {article} } Rail track geometry irregularities can lead to ride discomfort for passengers and redistribution of wheel loads potentially causing derailments. Current techniques for monitoring track settlements involve the use of discrete sensors or vehicle mounted sensors, which make it difficult to capture either spatial or temporal variations in settlement. Shape Array sensors (SAA) can potentially be used to capture temporal variations in distributed track settlement profiles to monitor and investigate potential track geometry irregularities and inform track maintenance programs. In this study, a site with known ground deformation issues (i.e. soft spots) was monitored with both an SAA and intermittent LIDAR scans. The objectives of the research were to investigate the accuracy of the SAA for measuring track settlements based on a comparison with LIDAR data, use those measurements to assess track irregularities at the site, evaluate temporal and seasonal changes in rail deformation, and to gain insight into the underlying causes of the ground deformation issues. The SAA was found to provide comparable settlement measurements to those from the LIDAR with the added advantage that the data could be used to assess settlement rates. At this particular site, long-term rail settlements were found to be a function of seasonal/climatic conditions with slower settlement rates in the winter and higher rates in the summer. In addition, the measurements indicated that the deformation issues are potentially caused by an asymmetric bearing failure. |
Sari, Y A; Kumral, M Clustering-based iterative approach to stope layout optimization for sublevel stoping Journal Article Journal of the Southern African Institute of Mining and Metallurgy, 121 , pp. 97 - 106, 2021, ISSN: 2225-6253. @article{SARI2021, title = {Clustering-based iterative approach to stope layout optimization for sublevel stoping}, author = {Y A Sari and M Kumral}, url = {http://www.scielo.org.za/scielo.php?script=sci_arttext&pid=S2225-62532021000300003&nrm=iso}, issn = {2225-6253}, year = {2021}, date = {2021-03-01}, journal = {Journal of the Southern African Institute of Mining and Metallurgy}, volume = {121}, pages = {97 - 106}, publisher = {scieloza}, crossref = {10.17159/2411-9717/1237/2021}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Suman, Shashi; Etemad, Ali; Rivest, Francois Potential Impacts of Smart Homes on Human Behavior: A Reinforcement Learning Approach Miscellaneous 2021. BibTeX | Tags: @misc{suman2021potential, title = {Potential Impacts of Smart Homes on Human Behavior: A Reinforcement Learning Approach}, author = {Shashi Suman and Ali Etemad and Francois Rivest}, year = {2021}, date = {2021-02-26}, keywords = {}, pubstate = {published}, tppubtype = {misc} } |
Lawal, Ghani; Greenspan, Michael Procam Calibration from a Single Pose of a Planar Target Journal Article Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, 2021, ISBN: 9789897584886. @article{Lawal_2021, title = {Procam Calibration from a Single Pose of a Planar Target}, author = {Ghani Lawal and Michael Greenspan}, url = {http://dx.doi.org/10.5220/0010327708170827}, doi = {10.5220/0010327708170827}, isbn = {9789897584886}, year = {2021}, date = {2021-02-22}, journal = {Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications}, publisher = {SCITEPRESS - Science and Technology Publications}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Sepas-Moghaddam, Alireza; Etemad, Ali Deep Gait Recognition: A Survey Miscellaneous 2021. BibTeX | Tags: @misc{sepasmoghaddam2021deep, title = {Deep Gait Recognition: A Survey}, author = {Alireza Sepas-Moghaddam and Ali Etemad}, year = {2021}, date = {2021-02-18}, keywords = {}, pubstate = {published}, tppubtype = {misc} } |
Mokogwu, C N; Hashtrudi-Zaad, K Platoon String Stability: A Passivity Perspective Inproceedings 2020 IEEE 3rd Connected and Automated Vehicles Symposium (CAVS), pp. 1-6, 2021. @inproceedings{9334649, title = {Platoon String Stability: A Passivity Perspective}, author = {C N Mokogwu and K Hashtrudi-Zaad}, doi = {10.1109/CAVS51000.2020.9334649}, year = {2021}, date = {2021-02-01}, booktitle = {2020 IEEE 3rd Connected and Automated Vehicles Symposium (CAVS)}, pages = {1-6}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } |
Woods, Joshua E; Yang, Yuan-Sen; Chen, Pei-Ching; Lau, David T; Erochko, Jeffrey Automated Crack Detection and Damage Index Calculation for RC Structures Using Image Analysis and Fractal Dimension Journal Article Transportation Geotechnics, 147 (4), pp. 04021019, 2021. Abstract | Links | BibTeX | Tags: Civil Infrastructure, Image Analysis @article{doi:10.1061/(ASCE)ST.1943-541X.0002970, title = {Automated Crack Detection and Damage Index Calculation for RC Structures Using Image Analysis and Fractal Dimension}, author = {Joshua E. Woods and Yuan-Sen Yang and Pei-Ching Chen and David T. Lau and Jeffrey Erochko}, url = {https://doi.org/10.1061/(ASCE)ST.1943-541X.0002970}, doi = {10.1061/(ASCE)ST.1943-541X.0002970}, year = {2021}, date = {2021-01-29}, journal = {Transportation Geotechnics}, volume = {147}, number = {4}, pages = {04021019}, abstract = {Damage assessment in postearthquake reconnaissance of civil structures has traditionally relied upon the judgment of experienced site inspectors. In a step toward eliminating the need for on-site inspections, this paper presents a method for automatic postdisaster structural damage assessment of reinforced concrete (RC) structures. The method uses digital image correlation to automatically detect cracks in RC structural elements and correlates the analyzed crack distributions to a damage index using fractal dimension. The method is applied in this study to track the progression of damage in a planar RC shear wall and a cylindrical RC containment vessel tested in a laboratory under reversed cyclic loading. The results from the proposed damage index are compared with quantitative and qualitative damage indices that have been used in the past to evaluate damage levels in RC structures. The results demonstrate the ability of the method to measure crack distributions and automatically correlate them to a damage index. The method is also applied to an RC shear wall tested under an actual earthquake ground motion record using hybrid simulation to evaluate its performance in a more realistic damage assessment scenario. The results show that the method is able to track the progression of damage to an RC structural element in a realistic earthquake damage scenario. Based on the results, damage grades are proposed that can be used to relate the automatically computed damage index to a specific damage level. The results show great promise for the automatic damage assessment method and is a first step towards automated postdisaster damage assessment of RC structures using digital image correlation.}, keywords = {Civil Infrastructure, Image Analysis}, pubstate = {published}, tppubtype = {article} } Damage assessment in postearthquake reconnaissance of civil structures has traditionally relied upon the judgment of experienced site inspectors. In a step toward eliminating the need for on-site inspections, this paper presents a method for automatic postdisaster structural damage assessment of reinforced concrete (RC) structures. The method uses digital image correlation to automatically detect cracks in RC structural elements and correlates the analyzed crack distributions to a damage index using fractal dimension. The method is applied in this study to track the progression of damage in a planar RC shear wall and a cylindrical RC containment vessel tested in a laboratory under reversed cyclic loading. The results from the proposed damage index are compared with quantitative and qualitative damage indices that have been used in the past to evaluate damage levels in RC structures. The results demonstrate the ability of the method to measure crack distributions and automatically correlate them to a damage index. The method is also applied to an RC shear wall tested under an actual earthquake ground motion record using hybrid simulation to evaluate its performance in a more realistic damage assessment scenario. The results show that the method is able to track the progression of damage to an RC structural element in a realistic earthquake damage scenario. Based on the results, damage grades are proposed that can be used to relate the automatically computed damage index to a specific damage level. The results show great promise for the automatic damage assessment method and is a first step towards automated postdisaster damage assessment of RC structures using digital image correlation. |
Sepas-Moghaddam, Alireza; Etemad, Ali; Pereira, Fernando; Correia, Paulo Lobato CapsField: Light Field-Based Face and Expression Recognition in the Wild Using Capsule Routing Journal Article IEEE Transactions on Image Processing, 30 , pp. 2627–2642, 2021, ISSN: 1941-0042. @article{Sepas_Moghaddam_2021, title = {CapsField: Light Field-Based Face and Expression Recognition in the Wild Using Capsule Routing}, author = {Alireza Sepas-Moghaddam and Ali Etemad and Fernando Pereira and Paulo Lobato Correia}, url = {http://dx.doi.org/10.1109/TIP.2021.3054476}, doi = {10.1109/tip.2021.3054476}, issn = {1941-0042}, year = {2021}, date = {2021-01-10}, journal = {IEEE Transactions on Image Processing}, volume = {30}, pages = {2627–2642}, publisher = {Institute of Electrical and Electronics Engineers (IEEE)}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Ahmed, Mirza Tahir; Ziauddin, Sheikh; Marshall, Joshua; Greenspan, Michael Point Cloud Registration Using Virtual Interest Points from Macaulay’s Resultant of Quadric Surfaces Journal Article Journal of Mathematical Imaging and Vision, 2021. @article{article, title = {Point Cloud Registration Using Virtual Interest Points from Macaulay’s Resultant of Quadric Surfaces}, author = {Mirza Tahir Ahmed and Sheikh Ziauddin and Joshua Marshall and Michael Greenspan}, doi = {10.1007/s10851-020-01013-z}, year = {2021}, date = {2021-01-07}, journal = {Journal of Mathematical Imaging and Vision}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Hui, Binyuan; Geng, Ruiying; Ren, Qiyu; Li, Binhua; Li, Yongbin; Sun, Jian; Huang, Fei; Si, Luo; Zhu, Pengfei; Zhu, Xiaodan Dynamic Hybrid Relation Network for Cross-Domain Context-Dependent Semantic Parsing Miscellaneous Forthcoming Forthcoming. @misc{hui2021dynamic, title = {Dynamic Hybrid Relation Network for Cross-Domain Context-Dependent Semantic Parsing}, author = {Binyuan Hui and Ruiying Geng and Qiyu Ren and Binhua Li and Yongbin Li and Jian Sun and Fei Huang and Luo Si and Pengfei Zhu and Xiaodan Zhu}, url = {https://arxiv.org/pdf/2101.01686.pdf }, year = {2021}, date = {2021-01-05}, keywords = {}, pubstate = {forthcoming}, tppubtype = {misc} } |
Uppal, Hardik; Sepas-Moghaddam, Alireza; Greenspan, Michael; Etemad, Ali Depth as Attention for Face Representation Learning Journal Article IEEE Transactions on Information Forensics and Security, 16 , pp. 2461–2476, 2021, ISSN: 1556-6021. @article{Uppal_2021, title = {Depth as Attention for Face Representation Learning}, author = {Hardik Uppal and Alireza Sepas-Moghaddam and Michael Greenspan and Ali Etemad}, url = {http://dx.doi.org/10.1109/TIFS.2021.3053458}, doi = {10.1109/tifs.2021.3053458}, issn = {1556-6021}, year = {2021}, date = {2021-01-03}, journal = {IEEE Transactions on Information Forensics and Security}, volume = {16}, pages = {2461–2476}, publisher = {Institute of Electrical and Electronics Engineers (IEEE)}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Sarkar, Pritam; Etemad, Ali CardioGAN: Attentive Generative Adversarial Network with Dual Discriminators for Synthesis of ECG from PPG Inproceedings AAAI Conference on Artificial Intelligence, 2021. BibTeX | Tags: @inproceedings{sarkar2020cardiogan, title = {CardioGAN: Attentive Generative Adversarial Network with Dual Discriminators for Synthesis of ECG from PPG}, author = {Pritam Sarkar and Ali Etemad}, year = {2021}, date = {2021-01-01}, publisher = {AAAI Conference on Artificial Intelligence}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } |
Davoodnia, Vandad; Ghorbani, Saeed; Etemad, Ali In-bed Pressure-based Pose Estimation using Image Space Representation Learning Inproceedings International Conference on Acoustics, Speech, and Signal Processing (ICASSP)", 2021. BibTeX | Tags: @inproceedings{davoodnia2020inbed, title = {In-bed Pressure-based Pose Estimation using Image Space Representation Learning}, author = {Vandad Davoodnia and Saeed Ghorbani and Ali Etemad}, year = {2021}, date = {2021-01-01}, publisher = {International Conference on Acoustics, Speech, and Signal Processing (ICASSP)"}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } |
de Araujo, Arthur Cruz; Etemad, Ali End-to-End Prediction of Parcel Delivery Time with Deep Learning for Smart-City Applications Miscellaneous 2021, (IEEE Internet of Things Journal). BibTeX | Tags: @misc{dearaujo2020endtoend, title = {End-to-End Prediction of Parcel Delivery Time with Deep Learning for Smart-City Applications}, author = {Arthur Cruz de Araujo and Ali Etemad}, year = {2021}, date = {2021-01-01}, note = {IEEE Internet of Things Journal}, keywords = {}, pubstate = {published}, tppubtype = {misc} } |
2020 |
Tan, Chao-Hong; Yang, Xiaoyu; Zheng, Zióu; Li, Tianda; Feng, Yufei; Gu, Jia-Chen; Liu, Quan; Liu, Dan; Ling, Zhen-Hua; Zhu, Xiaodan Learning to Retrieve Entity-Aware Knowledge and Generate Responses with Copy Mechanism for Task-Oriented Dialogue Systems Miscellaneous 2020. @misc{tan2020learning, title = {Learning to Retrieve Entity-Aware Knowledge and Generate Responses with Copy Mechanism for Task-Oriented Dialogue Systems}, author = {Chao-Hong Tan and Xiaoyu Yang and Zióu Zheng and Tianda Li and Yufei Feng and Jia-Chen Gu and Quan Liu and Dan Liu and Zhen-Hua Ling and Xiaodan Zhu}, url = {https://arxiv.org/pdf/2012.11937.pdf}, year = {2020}, date = {2020-12-22}, keywords = {}, pubstate = {published}, tppubtype = {misc} } |
Vuuren, Jansen-van R D; Nunzi, J -M; Givigi, S N Frontiers in Photosensor Materials and Designs for New Image Sensor Applications Journal Article IEEE Sensors Journal, pp. 1-1, 2020. @article{9286498, title = {Frontiers in Photosensor Materials and Designs for New Image Sensor Applications}, author = {R D Jansen-van Vuuren and J -M Nunzi and S N Givigi}, doi = {10.1109/JSEN.2020.3043288}, year = {2020}, date = {2020-12-08}, journal = {IEEE Sensors Journal}, pages = {1-1}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Woods, Joshua E; Lau, David T; Erochko, Jeffrey Evaluation by Hybrid Simulation of Earthquake-Damaged RC Walls Repaired for In-Plane Bending with Single-Sided CFRP Sheets Journal Article Journal of Composites for Construction, 24 (6), pp. 04020073, 2020. Abstract | Links | BibTeX | Tags: @article{doi:10.1061/(ASCE)CC.1943-5614.0001085, title = {Evaluation by Hybrid Simulation of Earthquake-Damaged RC Walls Repaired for In-Plane Bending with Single-Sided CFRP Sheets}, author = {Joshua E Woods and David T Lau and Jeffrey Erochko}, url = {https://ascelibrary.org/doi/abs/10.1061/%28ASCE%29CC.1943-5614.0001085}, doi = {10.1061/(ASCE)CC.1943-5614.0001085}, year = {2020}, date = {2020-12-01}, journal = {Journal of Composites for Construction}, volume = {24}, number = {6}, pages = {04020073}, abstract = {The realistic seismic response of two damaged reinforced concrete (RC) shear walls repaired using externally bonded carbon fiber-reinforced polymer (CFRP) sheets was evaluated using hybrid simulation. The CFRP repair used horizontal and vertical CFRP layers applied to a single side of the wall that were anchored with a steel tube anchor system and CFRP fan anchors. The objective of the CFRP repair was to restore the initial stiffness and restore or increase the strength, ductility, and energy dissipation capacity of the damaged walls. Hybrid simulation was used to evaluate the effectiveness of the repair strategy under real earthquake ground motion records with realistic boundary conditions, including the effects of axial load, shear force, and overturning moment. The results show that the single-sided application of the CFRP sheets restored the seismic performance of the damaged RC shear walls tested in this study. Hybrid simulation is shown to be an efficient test method to experimentally study the seismic response of a structural component with realistic boundary conditions over a range of earthquake hazard levels.}, keywords = {}, pubstate = {published}, tppubtype = {article} } The realistic seismic response of two damaged reinforced concrete (RC) shear walls repaired using externally bonded carbon fiber-reinforced polymer (CFRP) sheets was evaluated using hybrid simulation. The CFRP repair used horizontal and vertical CFRP layers applied to a single side of the wall that were anchored with a steel tube anchor system and CFRP fan anchors. The objective of the CFRP repair was to restore the initial stiffness and restore or increase the strength, ductility, and energy dissipation capacity of the damaged walls. Hybrid simulation was used to evaluate the effectiveness of the repair strategy under real earthquake ground motion records with realistic boundary conditions, including the effects of axial load, shear force, and overturning moment. The results show that the single-sided application of the CFRP sheets restored the seismic performance of the damaged RC shear walls tested in this study. Hybrid simulation is shown to be an efficient test method to experimentally study the seismic response of a structural component with realistic boundary conditions over a range of earthquake hazard levels. |
Li, Karly Kudrinko; Emile Flavin; Xiaodan Zhu; Qingguo Wearable Sensor-Based Sign Language Recognition: A Comprehensive Review Journal Article IEEE Reviews in Biomedical Engineering, pp. 1-1, 2020, ISSN: 1941-1189. Abstract | Links | BibTeX | Tags: Gesture recognition;Assistive technology;Cameras;Auditory system;Three-dimensional displays;Sociology;Statistics;wearable sensors;sign language;gesture recognition;machine learning;sensor systems @article{9178440, title = {Wearable Sensor-Based Sign Language Recognition: A Comprehensive Review}, author = {Karly Kudrinko; Emile Flavin; Xiaodan Zhu; Qingguo Li}, url = {https://arxiv.org/pdf/2101.01686.pdf}, doi = {10.1109/RBME.2020.3019769}, issn = {1941-1189}, year = {2020}, date = {2020-11-26}, journal = {IEEE Reviews in Biomedical Engineering}, pages = {1-1}, abstract = {Sign language is used as a primary form of communication by many people who are Deaf, deafened, hard of hearing, and non-verbal. Communication barriers exist for members of these populations during daily interactions with those who are unable to understand or use sign language. Advancements in technology and machine learning techniques have led to the development of innovative approaches for gesture recognition. This literature review focuses on analyzing studies that use wearable sensor-based systems to classify sign language gestures. A review of 72 studies from 1991 to 2019 was performed to identify trends, best practices, and common challenges. Attributes including sign language variation, sensor configuration, classification method, study design, and performance metrics were analyzed and compared. Results from this literature review could aid in the development of user-centred and robust wearable sensor-based systems for sign language recognition.}, keywords = {Gesture recognition;Assistive technology;Cameras;Auditory system;Three-dimensional displays;Sociology;Statistics;wearable sensors;sign language;gesture recognition;machine learning;sensor systems}, pubstate = {published}, tppubtype = {article} } Sign language is used as a primary form of communication by many people who are Deaf, deafened, hard of hearing, and non-verbal. Communication barriers exist for members of these populations during daily interactions with those who are unable to understand or use sign language. Advancements in technology and machine learning techniques have led to the development of innovative approaches for gesture recognition. This literature review focuses on analyzing studies that use wearable sensor-based systems to classify sign language gestures. A review of 72 studies from 1991 to 2019 was performed to identify trends, best practices, and common challenges. Attributes including sign language variation, sensor configuration, classification method, study design, and performance metrics were analyzed and compared. Results from this literature review could aid in the development of user-centred and robust wearable sensor-based systems for sign language recognition. |
Kudrinko, Karly; Flavin, Emile; Zhu, Xiaodan; Li, Qingguo Wearable Sensor-Based Sign Language Recognition: A Comprehensive Review Journal Article IEEE Reviews in Biomedical Engineering, 14 , pp. 82-97, 2020. @article{9178440b, title = {Wearable Sensor-Based Sign Language Recognition: A Comprehensive Review}, author = {Karly Kudrinko and Emile Flavin and Xiaodan Zhu and Qingguo Li}, doi = {10.1109/RBME.2020.3019769}, year = {2020}, date = {2020-11-26}, journal = {IEEE Reviews in Biomedical Engineering}, volume = {14}, pages = {82-97}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Jing, Guoqing; Siahkouhi, Mohammad; Edwards, Riley J; Dersch, Marcus S; Hoult, N A Smart railway sleepers - a review of recent developments, challenges, and future prospects Journal Article Construction and Building Materials, pp. 121533, 2020, ISSN: 0950-0618. Abstract | Links | BibTeX | Tags: Concrete sleeper, High speed railway, Self-sensing, Smart railway sleeper, Structural health monitoring, Sustainability design @article{JING2020121533, title = {Smart railway sleepers - a review of recent developments, challenges, and future prospects}, author = {Guoqing Jing and Mohammad Siahkouhi and J Riley Edwards and Marcus S Dersch and N A Hoult}, url = {http://www.sciencedirect.com/science/article/pii/S0950061820335376}, doi = {https://doi.org/10.1016/j.conbuildmat.2020.121533}, issn = {0950-0618}, year = {2020}, date = {2020-11-20}, journal = {Construction and Building Materials}, pages = {121533}, abstract = {To date, the application of self-sensing or other structural health monitoring (SHM) technologies in railway sleepers is limited. Development and deployment of smart sleepers can provide value to rail infrastructure owners and maintainers by informing maintenance and replacement decisions, collecting support condition data providing insight into the health of the track substructure, supporting numerical model development and validation, and providing energy harvesting capabilities. These result in extending the service life of sleepers and the surrounding railway infrastructure providing a key element of sustainable design for the track systems. This paper provides a comprehensive review of recent smart sleeper developments and describes challenges to more widespread adoption. Limitations to using smart sleepers include their production cost and limited data on long term performance, which can be incompatible with the sleeper’s service life. Potential solutions for overcoming the challenges associated with the application of smart sleeper technologies include the use of intrinsic self-sensing concrete, adding self-healing features, taking advantage of recent wireless sensing developments, and connecting with the emerging use of Internet of Things (IoT) technology. Opportunities abound to expand smart sleeper deployment for condition monitoring and real-time management of track assets, to decrease the life cycle cost (LCC) of track components and increase infrastructure availability whereby increasing track capacity.}, keywords = {Concrete sleeper, High speed railway, Self-sensing, Smart railway sleeper, Structural health monitoring, Sustainability design}, pubstate = {published}, tppubtype = {article} } To date, the application of self-sensing or other structural health monitoring (SHM) technologies in railway sleepers is limited. Development and deployment of smart sleepers can provide value to rail infrastructure owners and maintainers by informing maintenance and replacement decisions, collecting support condition data providing insight into the health of the track substructure, supporting numerical model development and validation, and providing energy harvesting capabilities. These result in extending the service life of sleepers and the surrounding railway infrastructure providing a key element of sustainable design for the track systems. This paper provides a comprehensive review of recent smart sleeper developments and describes challenges to more widespread adoption. Limitations to using smart sleepers include their production cost and limited data on long term performance, which can be incompatible with the sleeper’s service life. Potential solutions for overcoming the challenges associated with the application of smart sleeper technologies include the use of intrinsic self-sensing concrete, adding self-healing features, taking advantage of recent wireless sensing developments, and connecting with the emerging use of Internet of Things (IoT) technology. Opportunities abound to expand smart sleeper deployment for condition monitoring and real-time management of track assets, to decrease the life cycle cost (LCC) of track components and increase infrastructure availability whereby increasing track capacity. |
Alghamdi, Ahmed; Chan, Wai-Yip Modified ESTOI for improving speech intelligibility prediction Inproceedings 2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), pp. 1-5, 2020. @inproceedings{9255677, title = {Modified ESTOI for improving speech intelligibility prediction}, author = {Ahmed Alghamdi and Wai-Yip Chan}, doi = {10.1109/CCECE47787.2020.9255677}, year = {2020}, date = {2020-11-19}, booktitle = {2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE)}, pages = {1-5}, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } |
Li, Xiaoyan; Kiringa, Iluju; Yeap, Tet; Zhu, Xiaodan; Li, Yifeng Anomaly detection based on unsupervised disentangled representation learning in combination with manifold learning Journal Article International Joint Conference on Neural Networks (IJCNN), 2020. Abstract | Links | BibTeX | Tags: Artificial Intelligence, Natural Language Processing @article{li2019anomaly, title = {Anomaly detection based on unsupervised disentangled representation learning in combination with manifold learning}, author = { Xiaoyan Li and Iluju Kiringa and Tet Yeap and Xiaodan Zhu and Yifeng Li}, url = {https://openreview.net/forum?id=r1xHxgrKwr}, year = {2020}, date = {2020-11-05}, journal = {International Joint Conference on Neural Networks (IJCNN)}, abstract = {Identifying anomalous samples from highly complex and unstructured data is a crucial but challenging task in a variety of intelligent systems. In this paper, we present a novel deep anomaly detection framework named AnoDM (standing for Anomaly detection based on unsupervised Disentangled representation learning and Manifold learning). The disentanglement learning is currently implemented by beta-VAE for automatically discovering interpretable factorized latent representations in a completely unsupervised manner. The manifold learning is realized by t-SNE for projecting the latent representations to a 2D map. We define a new anomaly score function by combining beta-VAE's reconstruction error in the raw feature space and local density estimation in the t-SNE space. AnoDM was evaluated on both image and time-series data and achieved better results than models that use just one of the two measures and other deep learning methods.}, keywords = {Artificial Intelligence, Natural Language Processing}, pubstate = {published}, tppubtype = {article} } Identifying anomalous samples from highly complex and unstructured data is a crucial but challenging task in a variety of intelligent systems. In this paper, we present a novel deep anomaly detection framework named AnoDM (standing for Anomaly detection based on unsupervised Disentangled representation learning and Manifold learning). The disentanglement learning is currently implemented by beta-VAE for automatically discovering interpretable factorized latent representations in a completely unsupervised manner. The manifold learning is realized by t-SNE for projecting the latent representations to a 2D map. We define a new anomaly score function by combining beta-VAE's reconstruction error in the raw feature space and local density estimation in the t-SNE space. AnoDM was evaluated on both image and time-series data and achieved better results than models that use just one of the two measures and other deep learning methods. |
Sepas-Moghaddam, Alireza; Etemad, Ali; Pereira, Fernando; Correia, Paulo Lobato Long Short-Term Memory With Gate and State Level Fusion for Light Field-Based Face Recognition Journal Article IEEE Transactions on Information Forensics and Security, 16 , pp. 1365-1379, 2020. @article{9249026, title = {Long Short-Term Memory With Gate and State Level Fusion for Light Field-Based Face Recognition}, author = {Alireza Sepas-Moghaddam and Ali Etemad and Fernando Pereira and Paulo Lobato Correia}, doi = {10.1109/TIFS.2020.3036242}, year = {2020}, date = {2020-11-05}, journal = {IEEE Transactions on Information Forensics and Security}, volume = {16}, pages = {1365-1379}, keywords = {}, pubstate = {published}, tppubtype = {article} } |
Yang, Xiaoyu; Nie, Feng; Feng, Yufei; Liu, Quan; Chen, Zhigang; Zhu, Xiaodan Program Enhanced Fact Verification with Verbalization and Graph Attention Network Inproceedings EMNLP, 2020, (arXiv.org > cs > arXiv:2010.03084v4 ). Abstract | Links | BibTeX | Tags: @inproceedings{yang2020program, title = {Program Enhanced Fact Verification with Verbalization and Graph Attention Network}, author = {Xiaoyu Yang and Feng Nie and Yufei Feng and Quan Liu and Zhigang Chen and Xiaodan Zhu}, url = {https://arxiv.org/pdf/2010.03084.pdf}, year = {2020}, date = {2020-11-04}, booktitle = {EMNLP}, abstract = {Performing fact verification based on structured data is important for many real-life applications and is a challenging research problem, particularly when it involves both symbolic operations and informal inference based on language understanding. In this paper, we present a Program-enhanced Verbalization and Graph Attention Network (ProgVGAT) to integrate programs and execution into textual inference models. Specifically, a verbalization with program execution model is proposed to accumulate evidences that are embedded in operations over the tables. Built on that, we construct the graph attention verification networks, which are designed to fuse different sources of evidences from verbalized program execution, program structures, and the original statements and tables, to make the final verification decision. To support the above framework, we propose a program selection module optimized with a new training strategy based on margin loss, to produce more accurate programs, which is shown to be effective in enhancing the final verification results. Experimental results show that the proposed framework achieves the new state-of-the-art performance, a 74.4% accuracy, on the benchmark dataset TABFACT. }, note = {arXiv.org > cs > arXiv:2010.03084v4 }, keywords = {}, pubstate = {published}, tppubtype = {inproceedings} } Performing fact verification based on structured data is important for many real-life applications and is a challenging research problem, particularly when it involves both symbolic operations and informal inference based on language understanding. In this paper, we present a Program-enhanced Verbalization and Graph Attention Network (ProgVGAT) to integrate programs and execution into textual inference models. Specifically, a verbalization with program execution model is proposed to accumulate evidences that are embedded in operations over the tables. Built on that, we construct the graph attention verification networks, which are designed to fuse different sources of evidences from verbalized program execution, program structures, and the original statements and tables, to make the final verification decision. To support the above framework, we propose a program selection module optimized with a new training strategy based on margin loss, to produce more accurate programs, which is shown to be effective in enhancing the final verification results. Experimental results show that the proposed framework achieves the new state-of-the-art performance, a 74.4% accuracy, on the benchmark dataset TABFACT. |
2021 |
Speech Intelligibility Prediction Using Spectro-Temporal Modulation Analysis Journal Article IEEE/ACM Transactions on Audio, Speech, and Language Processing, 29 , pp. 210-225, 2021. |
Drag force and reconfiguration of cultivated Saccharina latissima in current Journal Article Aquacultural Engineering, 94 , pp. 102169, 2021, ISSN: 0144-8609. |
A hybrid position–rate teleoperation system Journal Article Robotics and Autonomous Systems, 141 , pp. 103781, 2021, ISSN: 0921-8890. |
Experimental Validation of a Steering Control System using an Adaptive Fuzzy Controller and Computer Vision Inproceedings 2021 IEEE International Systems Conference (SysCon), pp. 1-8, 2021. |
A System For P300 Detection Applied To Vehicle Navigation Inproceedings 2021 IEEE International Systems Conference (SysCon), pp. 1-8, 2021. |
Systems Design for EEG Signal Classification of Sensorimotor Activity Using Machine Learning Inproceedings 2021 IEEE International Systems Conference (SysCon), pp. 1-7, 2021. |
The Management of Risks and Benefits When Engineering Interactive Digital Systems Inproceedings Companion of the 2021 ACM SIGCHI Symposium on Engineering Interactive Computing Systems, pp. 81–83, Association for Computing Machinery, Virtual Event, Netherlands, 2021, ISBN: 9781450384490. |
Estimating pose from pressure data for smart beds with deep image-based pose estimators Journal Article Applied Intelligence, 2021, ISBN: Davoodnia2021. |
Flow-induced cross-flow vibrations of long flexible cylinder with an upstream wake interference Journal Article Physics of Fluids, 33 (6), pp. 065104, 2021. |
Preview, Attend and Review: Schema-Aware Curriculum Learning for Multi-Domain Dialog State Tracking Miscellaneous 2021. |
Removing energy with an exoskeleton reduces the metabolic cost of walking Journal Article Science, 372 (6545), pp. 957–960, 2021, ISSN: 0036-8075. |
CONTEXT BASED PREDICTION OF A COMMUNICATION TIME Journal Article (20210150381), 2021. |
Partner Matters! An Empirical Study on Fusing Personas for Personalized Response Selection in Retrieval-Based Chatbots Miscellaneous 2021. |
Location Privacy-Preserving Task Recommendation with Geometric Range Query in Mobile Crowdsensing Journal Article IEEE Transactions on Mobile Computing, pp. 1-1, 2021. |
Have You Made a Decision? Where? A Pilot Study on Interpretability of Polarity Analysis Based on Advising Problem Inproceedings ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 6928-6932, 2021. |
Multi-Perspective LSTM for Joint Visual Representation Learning Miscellaneous 2021. |
Evaluating the behaviour of functionally graded reinforced concrete without transverse reinforcement using distributed sensing Journal Article Construction and Building Materials, 295 , pp. 123612, 2021, ISSN: 0950-0618. |
Oriented Bounding Boxes for Small and Freely Rotated Objects Miscellaneous 2021. |
Flow-based Autoregressive Structured Prediction of Human Motion Miscellaneous 2021. |
IEEE Transactions on Vehicular Technology, pp. 1-1, 2021. |
Vote from the Center: 6 DoF Pose Estimation in RGB-D Images by Radial Keypoint Voting Miscellaneous 2021. |
Teacher-Student Adversarial Depth Hallucination to Improve Face Recognition Miscellaneous 2021. |
Deep Learning-Based Ground Vibration Monitoring: Reinforcement Learning and RNN-CNN Approach Journal Article IEEE Geoscience and Remote Sensing Letters, pp. 1-5, 2021. |
Communications design in multi-robotic systems Inproceedings 2020 International Conference on Communications, Signal Processing, and their Applications (ICCSPA), pp. 1-6, 2021. |
Deep Reinforcement Learning-based Task Offloading Decision in the Time Varying Channel Inproceedings 2021 International Conference on Electronics, Information, and Communication (ICEIC), pp. 1-4, 2021. |
Improving Text-to-SQL with Schema Dependency Learning Online 2021. |
Evaluation of Shape Array sensors to quantify the spatial distribution and seasonal rate of track settlement Journal Article Transportation Geotechnics, 27 , pp. 100487, 2021, ISSN: 2214-3912. |
Clustering-based iterative approach to stope layout optimization for sublevel stoping Journal Article Journal of the Southern African Institute of Mining and Metallurgy, 121 , pp. 97 - 106, 2021, ISSN: 2225-6253. |
Potential Impacts of Smart Homes on Human Behavior: A Reinforcement Learning Approach Miscellaneous 2021. |
Procam Calibration from a Single Pose of a Planar Target Journal Article Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, 2021, ISBN: 9789897584886. |
Deep Gait Recognition: A Survey Miscellaneous 2021. |
Platoon String Stability: A Passivity Perspective Inproceedings 2020 IEEE 3rd Connected and Automated Vehicles Symposium (CAVS), pp. 1-6, 2021. |
Automated Crack Detection and Damage Index Calculation for RC Structures Using Image Analysis and Fractal Dimension Journal Article Transportation Geotechnics, 147 (4), pp. 04021019, 2021. |
CapsField: Light Field-Based Face and Expression Recognition in the Wild Using Capsule Routing Journal Article IEEE Transactions on Image Processing, 30 , pp. 2627–2642, 2021, ISSN: 1941-0042. |
Point Cloud Registration Using Virtual Interest Points from Macaulay’s Resultant of Quadric Surfaces Journal Article Journal of Mathematical Imaging and Vision, 2021. |
Dynamic Hybrid Relation Network for Cross-Domain Context-Dependent Semantic Parsing Miscellaneous Forthcoming Forthcoming. |
Depth as Attention for Face Representation Learning Journal Article IEEE Transactions on Information Forensics and Security, 16 , pp. 2461–2476, 2021, ISSN: 1556-6021. |
CardioGAN: Attentive Generative Adversarial Network with Dual Discriminators for Synthesis of ECG from PPG Inproceedings AAAI Conference on Artificial Intelligence, 2021. |
In-bed Pressure-based Pose Estimation using Image Space Representation Learning Inproceedings International Conference on Acoustics, Speech, and Signal Processing (ICASSP)", 2021. |
End-to-End Prediction of Parcel Delivery Time with Deep Learning for Smart-City Applications Miscellaneous 2021, (IEEE Internet of Things Journal). |
2020 |
Learning to Retrieve Entity-Aware Knowledge and Generate Responses with Copy Mechanism for Task-Oriented Dialogue Systems Miscellaneous 2020. |
Frontiers in Photosensor Materials and Designs for New Image Sensor Applications Journal Article IEEE Sensors Journal, pp. 1-1, 2020. |
Evaluation by Hybrid Simulation of Earthquake-Damaged RC Walls Repaired for In-Plane Bending with Single-Sided CFRP Sheets Journal Article Journal of Composites for Construction, 24 (6), pp. 04020073, 2020. |
Wearable Sensor-Based Sign Language Recognition: A Comprehensive Review Journal Article IEEE Reviews in Biomedical Engineering, pp. 1-1, 2020, ISSN: 1941-1189. |
Wearable Sensor-Based Sign Language Recognition: A Comprehensive Review Journal Article IEEE Reviews in Biomedical Engineering, 14 , pp. 82-97, 2020. |
Smart railway sleepers - a review of recent developments, challenges, and future prospects Journal Article Construction and Building Materials, pp. 121533, 2020, ISSN: 0950-0618. |
Modified ESTOI for improving speech intelligibility prediction Inproceedings 2020 IEEE Canadian Conference on Electrical and Computer Engineering (CCECE), pp. 1-5, 2020. |
Anomaly detection based on unsupervised disentangled representation learning in combination with manifold learning Journal Article International Joint Conference on Neural Networks (IJCNN), 2020. |
Long Short-Term Memory With Gate and State Level Fusion for Light Field-Based Face Recognition Journal Article IEEE Transactions on Information Forensics and Security, 16 , pp. 1365-1379, 2020. |
Program Enhanced Fact Verification with Verbalization and Graph Attention Network Inproceedings EMNLP, 2020, (arXiv.org > cs > arXiv:2010.03084v4 ). |