The College of Engineering and Computer Science

Permanent URI for this community

https://vinspace.online/handle/123456789/3

Education to create a generation of young leaders with the qualities, skills and knowledge needed to lead the development of science and technology, to increase productivity, and bring about radical innovation for the benefit of the society.

Browse

Now showing 1 - 20 of 65

Topic analysis in news via sparse learning: a case study on the 2016 US presidential elections
(2017-06) Ghaoui, Laurent El; Calafiore, Giuseppe C
Textual data such as tweets and news is abundant on the web. However, extracting useful information from such a deluge of data is hardly possible for a human. In this paper, we discuss automated text analysis methods based on sparse optimization. In particular, we use sparse PCA and Elastic Net regression for extracting intelligible topics from a big textual corpus and for obtaining time-based signals quantifying the strength of each topic in time. These signals can then be used as regressors for modeling or predicting other related numerical indices. We applied this setup to the analysis of the topics that arose during the 2016 US presidential elections, and we used the topic strength signals in order to model their influence on the election polls.
A Quantitative Analysis of the Effect of Human Detection and Segmentation Quality in Person Re-identification Performance
(2019) Binh Nguyen; Quan Nguyen; Lan Le; Thuy Pham; Nam Pham
Person re-identification, a problem of person identity association across camera views at different locations and times, is the second step in two-steps system for automatic video surveillance: person detection, tracking and person reidentification. However, most of the reported person Re-ID methods deal with the human regions of interest (ROIs) which are extracted manually with well-aligned bounding boxes. They mainly focus on designing discriminative feature descriptors and relevant metric learning on these manually-cropped human ROIs. This paper aims to answer two questions: (1) Do human detection and segmentation affect the performance of person reidentification?; (2) How to overcome the effect of human detection and segmentation with the state-of-the-art method for person re-identification? To answer these two questions, quantitative evaluations have been performed for both single-shot and multishot scenarios of person re-identification. Different state-of-the-art methods for human detection and segmentation have been evaluated on two benchmark datasets (VIPeR and PRID2011). The obtained results allow us to give some suggestions for developing fully automatic video surveillance systems.
QoE-Aware Video Streaming over HTTP and Software Defined Networking
(2019) Pham, Hong Thinh; Nguyen, Thanh Dat; Pham, Ngoc Nam; Nguyen, Huu Thanh; Truong, Thu Huong
Due to the increase in video streaming traffic over the Internet, more innovative methods are in demand for improving both the Quality of Experience (QoE) of users and the Quality of Service (QoS) of providers. In recent years, HTTP Adaptive Streaming (HAS) has received significant attention from both industry and academia based on its impacts in the enhancement of media streaming services. However, HAS-alone cannot guarantee a seamless viewing experience, since this highly relies on the Network Operators’ infrastructure and evolving network conditions. Along with the development of future Internet infrastructure, SoftwareDefined Networking (SDN) has been researched and newly implemented as a promising solution in improving the services of different Internet layers. In order to enhance the quality of video delivery, we try to combine the above two technologies, which has not been well-studied in academia. In this paper, we present a novel architecture incorporating bitrate adaptation and dynamic route allocation. At the client side, the adaptation logic of VBR videos streaming is built based on the MPEG-DASH standard. On the network side, an SDN controller is implemented with several routing strategies on top of the OpenFlow protocol. Our experimental results show that the proposed solution enhances at least 38% up to 185% in terms of average bitrate in comparison with some existing solutions as well as achieves a smoother viewing experience than the traditional Internet.
Scalable 360 Video Streaming using HTTP/2
(2019) Nguyen, Duc; Hoang, Van Trung; Hoang, Le Dieu Huong; Truong, Thu Huong; Pham, Ngoc Nam; Truong, Cong Thang
360-degree video is the main content type of Virtual Reality, providing users with an immersive viewing experience. In this paper, we propose a novel adaptation method for 360-degree video streaming over HTTP/2, which can provide a high viewing experience to users under time-varying network conditions and time-varying user head movements. The proposed method utilizes Scalable Video Coding to solve the trade-off between network adaptivity and user adaptivity. An optimal tile layer selection algorithm is provided. To cope with sudden throughput drops, the delivery of late layers is terminated using HTTP/2’s stream termination feature. Also, a tile layer updating scheme is proposed to deal with viewport estimation errors. Experimental results show that the proposed method can improve the average bitrate of viewport by 16-17% compared to a reference method.
SDN – based Dynamic Bandwidth Allocation for Multiple Video-on-Demand Players over HTTP
(2019) Pham, Hong Thinh; Pham, Ngoc Nam; Nguyen, Huu Thanh; Truong, Thu Huong
This nowadays, HTTP adaptive streaming (HAS) has been becoming the de-factor standard for video streaming over the multimedia network. However, HAS-alone cannot guarantee a seamless viewing experience, the adaptation of HAS in the management network faces a big challenge because of purely client-driven approaches which lead to unfair competition of available bandwidth resource when multiple bitrate-adaptive players share the same bottleneck network link. At that time, each HAS client tries independently to maximize its own bandwidth sharing, which leads to the competition of network resources among clients causing greatly reduced QoE (Quality of Experience) of end users. This competition will affect negatively several main metrics for each video player: fairness, efficiency, and stability. In this paper, we propose a scheme of bitrate adaptation for the HAS system combined with a Software Defined Networking (SDN)-based dynamic resource allocation, which aims to improve the quality of experience among competing clients. Our experiment results show that the proposed method significantly outperforms the conventional method on all the key QoE parameters under several scenarios.
An LSTM-based Approach for Overall Quality Prediction in HTTP Adaptive Streaming
(2019-04) Huyen Tran; Duc Nguyen; Nam Pham; Thang Truong; Duong Nguyen
HTTP Adaptive Streaming (HAS) has become a popular solution for multimedia delivery nowadays. In HAS, video quality is generally varying in each streaming session. Therefore, a key question in HTTP Adaptive Streaming is how to evaluate the overall quality of a streaming session. In this paper, we propose a machine-learning approach for overall quality prediction in HTTP Adaptive Streaming. In the proposed approach, each segment is represented by four features segment quality, stalling durations, content characteristics, and padding. The features are fed into a Long Short-Term Memory (LSTM) network that is capable of exploring temporal relations between segments. The overall quality of the streaming session is predicted from the outputs of the LSTM network using a linear regression module. Experiment results show that the proposed approach is effective in predicting the overall quality of streaming sessions. Also, it is found that our proposed approach outperforms four existing approaches.
A Subjective Study on User Perception Aspects in Virtual Reality
(2019-08-16) Huyen Tran; Nam Pham; Cuong Pham; Yong Ju Jung; Thang Truong
Three hundred and sixty degree video is becoming more and more popular on the Internet. By using a Head-Mounted Display, 360-degree video can render a Virtual Reality (VR) environment. However, it is still a big challenge to understand Quality of Experience (QoE) of 360-degree video since user experience during watching 360-degree video is a very complex phenomenon. In this paper, we aim to investigate four QoE aspects of 360-degree video, namely, perceptual quality, presence, cybersickness, and acceptability. In addition, four key QoE-affecting factors of encoding parameters, content motion, rendering device, and rendering mode are considered in our study. To the best of our knowledge, this is the first work that covers a large number of factors and QoE aspects of 360-degree video. In this study, a subjective experiment is conducted using 60 video versions generated from three original 360-degree videos. Based on statistical analysis of the obtained results, various findings on the impacts of the factors on the QoE aspects are provided. In particular, regarding the impacts of encoding parameters, it is found that the difference of QoE is negligible between video versions encoded at 4 K and 2.5 K resolutions. Also, it is suggested that 360-degree video should not be encoded at HD resolution or lower when watching in VR mode using Head Mounted Display. In addition, the bitrate for good QoE varies widely across different video contents. With respect to the content motion factor, its impact is statistically significant on the perceptual quality, presence, and cybersickness. In a comparison of two rendering device sets used in this study, there is no statistically significant difference found for the acceptability and cybersickness. However, the differences of the perceptual quality and presence are indicated to be statistically significant. Regarding the rendering mode, a comparison between VR and non-VR modes is also conducted. Although the non-VR mode always achieves higher perceptual quality scores and higher acceptability rates, more than half of the viewers prefer the VR mode to the non-VR mode when watching versions encoded at the resolutions of fHD or higher. By contrast, the non-VR mode is preferred at the HD resolution.
An Efficient QoE-Aware HTTP Adaptive Streaming over Software Defined Networking
(2020) Thinh Pham; Dat Nguyen; Nam Pham; Thanh Nguyen; Hien Nguyen; Huong Truong
Due to the increase in video streaming traffic over the Internet, more innovative methods are in demand for improving both Quality of Experience (QoE) of users and the Quality of Service (QoS) of providers. In recent years, HTTP Adaptive Streaming (HAS) has received significant attention from both industry and academia based on its impacts on the enhancement of media streaming services. However, HAS alone cannot guarantee a seamless viewing experience, since this highly relies on the Network Operators’ infrastructure and evolving network conditions. Along with the development of future Internet infrastructure, Software-Defined Networking (SDN) has been researched and newly implemented as a promising solution in improving services of different Internet layers. In this paper, we present a novel architecture incorporating bitrate adaptation and dynamic route allocation. On the client side, the adaptation logic of VBR video streaming is built based on the MPEGDASH standard. On the network side, an SDN controller is implemented with several routing strategies on top of the OpenFlow protocol. Our experimental results show that the proposed solution enhances at least 38% up to 185% in terms of average bitrate in comparison with some existing solutions as well as achieves a better viewing experience than the traditional Internet.
A Privacy-Preserving Framework for Surveillance Systems
(2020) Kok-Seng Wong; Tu Nguyen; Anuar Maratkhan; M. Fatih, Demirci
The ability to visually track people present in the scene is essential for any surveillance system. However, the widespread deployment and increased advancement of video surveillance systems have raised awareness of privacy to the public, i.e., human identity in the videos. The existing indoor surveillance systems allow people to be watched remotely and recorded continuously but do not prevent any party from viewing activities and collecting personal visual information of people in the videos. Because of this problem, we propose a privacy-preserving framework to provide each user (e.g., parents) with a personalized video where the user sees only selected target subjects (e.g., child, teacher, and intruder) while other faces are dynamically masked. The primary services in our framework consist of a video streaming service and a personalized service. The video streaming service is responsible for detecting, segmenting, recognizing, and masking face images of the human subjects in the video. Notably, it classifies human subjects into insider and outsider classes and then applies the de-identification (i.e., masking) to those in the insider class, including the target subjects. Subsequently, the personalized service receives the visual information (i.e., masked and unmasked faces) from the streaming service and processes it at the user’s mobile device. The output is then a personalized video for each user. For security reasons, we require surveillance videos stored in the cloud in an encrypted form. To ensure an individual remains anonymous in a group, we propose a dynamic masking approach to mask the human subjects in the video. Our framework can deliver both reliable visual privacy protection and video utility. For instance, users can have confidence that their target subjects are anonymized in other views. To utilize the personalized video, users can use analytics software installed on their mobile devices to analyze the activities of their target subjects.
Stability Properties of 1-Dimensional Hamiltonian Lattices with Non-analytic Potentials
(2020) Bountis, Anastasios; Kaloudis, Konstantinos; Oikonomou, Thomas; Manda, M. Bertin; Skokos, Charalampos
We investigate the local and global dynamics of two 1-Dimensional (1D) Hamiltonian lattices whose inter-particle forces are derived from non-analytic potentials. In particular, we study the dynamics of a model governed by a “graphene-type” force law and one inspired by Hollomon’s law describing “work-hardening” effects in certain elastic materials. Our main aim is to show that, although similarities with the analytic case exist, some of the local and global stability properties of non-analytic potentials are very different than those encountered in systems with polynomial interactions, as in the case of 1D Fermi-Pasta-Ulam-Tsingou (FPUT) lattices. Our approach is to study the motion in the neighborhood of simple periodic orbits representing continuations of normal modes of the corresponding linear system, as the number of particles N and the total energy E are increased. We find that the graphene-type model is remarkably stable up to escape energy levels where breakdown is expected, while the Hollomon lattice never breaks, yet is unstable at low energies and only attains stability at energies where the harmonic force becomes dominant. We suggest that, since our results hold for large N, it would be interesting to study analogous phenomena in the continuum limit where 1D lattices become strings.
Constrained crystal growth during solidification of particles and splats in uniform droplet sprays
(2020) Yiannos Ioannou; Hiroki Fukuda; Claus Rebholz; Yiliang Liao; Teiichi Ando; Charalabos C. Doumanidis
Uniform droplet spraying (UDS) is a novel process used to produce ideally narrow (mono-size) distributions of molten metal droplets for various applications. The crystallite size is a primary determinant of mechanical properties in solidified alloy deposits and thus in need of predictive modeling. This project reports on employing UDS as a paradigm for solidification modeling of mono-size solid droplets in an oil bath, as well as planar and globular splats on a cooling substrate, for magnesium alloys AZ91D and Mg97ZnY2. The model combines a nucleation and dendrite fragmentation description from solidification theory, with a framework for constrained growth of crystalline domains confined by adjacent developing ones. The latter is based on differential attributes of the dynamic temperature field during solidification, derived from semi-analytical expressions for the simple droplet and splat geometries above. The modeling results are validated against measured domain size distributions on section micrographs and found to be within a − 10% to + 14% estimation error range. Further improvement of the model via numerical thermal descriptions for off-line material design and optimization in additive manufacturing is discussed, along with its use as a real-time structural observer for closed-loop control based on temperature measurements in UDS-based processes.
FDM-Based 3D Printing of Polymer and Associated Composite: A Review on Mechanical Properties, Defects and Treatments
(2020) Wickramasinghe, Sachini; Truong Do; Phuong Tran
Fused deposition modeling (FDM) is one of the fastest-growing additive manufacturing methods used in printing fiber-reinforced composites (FRC). The performances of the resulting printed parts are limited compared to those by other manufacturing methods due to their inherent defects. Hence, the effort to develop treatment methods to overcome these drawbacks has accelerated during the past few years. The main focus of this study is to review the impact of those defects on the mechanical performance of FRC and therefore to discuss the available treatment methods to eliminate or minimize them in order to enhance the functional properties of the printed parts. As FRC is a combination of polymer matrix material and continuous or short reinforcing fibers, this review will thoroughly discuss both thermoplastic polymers and FRCs printed via FDM technology, including the effect of printing parameters such as layer thickness, infill pattern, raster angle and fibre orientation. The most common defects on printed parts, in particular, the void formation, surface roughness, and poor bonding between fibre and matrix, are explored. An inclusive discussion on the effectiveness of chemical, laser, heat, and ultrasound treatments to minimize these drawbacks is provided by this review.
Engineering a light–matter strong coupling regime in perovskite-based plasmonic metasurface: quasi-bound state in the continuum and exceptional points
(2020) Leran Lu; Quynh Le; Lydie Ferrier; Emmanuel Drouard; Christian Seassal; Son Nguyen
We present theoretically the formation of exciton–photon polaritons and exciton-surface plasmon polaritons in a perovskite-based subwavelength lattice on the metallic plane. It is shown that the later polaritons will be achieved as the perovskite layer is ultra-thin (<50 nm), while the co-existence of both polaritons will dominate, as the thickness of the perovskite metasurface approaches the wavelength scale. In the two cases, the lower polaritonic branches consist of dark and bright modes corresponding to infinite and finite radiative quality factors, respectively. Another salient property of this work is that it allows one to obtain exceptional points (EPs) in momentum space with a four-fold enhancement of the local density of states through engineering the perovskite metasurface. Our findings show that the perovskite metasurface is an attractive and rich platform to make polaritonic devices, even with the presence of a lossy metallic layer.
Input-Aware Dynamic Backdoor Attack
(2020) Anh Nguyen; Anh Tran
In recent years, neural backdoor attack has been considered to be a potential security threat to deep learning systems. Such systems, while achieving state-of-the-art performance on clean data, perform abnormally on inputs with predefined triggers. Current backdoor techniques, however, rely on uniform trigger patterns, which are easily detected and mitigated by current defense methods. In this work, we propose a novel backdoor attack technique in which the triggers vary from input to input. To achieve this goal, we implement an input-aware trigger generator driven by diversity loss. A novel cross-trigger test is applied to enforce trigger nonreusablity, making backdoor verification impossible. Experiments show that our method is efficient in various attack scenarios as well as multiple datasets. We further demonstrate that our backdoor can bypass state-of-the-art defense methods. An analysis with a famous neural network inspector again proves the stealthiness of the proposed attack. Our code is publicly available.
Multivariable control of ball-milled reactive material composition and structure
(2020) Matteo Aureli; Constantine C. Doumanidis; Aseel Gamal Suliman Hussien; Syed Murtaza Jaffar; Nikolaos Kostoglou; Yiliang Liao; Claus Rebholz; Charalabos C. Doumanidisf
In reactive bimetallic compounds such as Ni–Al multilayers, desirable thermo-kinetic properties upon ignition require simultaneously controlled geometric microstructure and material composition. This article establishes fundamental dynamical models of plastic deformation and material diffusion in ball milling processing of particulates from Ni and Al powders, for the purpose of designing and implementing feedback control strategies for process control. The role of heat dissipation from plastic yield and friction slip in affecting compressibility and diffusivity of the material is elucidated. The different sensitivity of compressibility and diffusivity to thermal power is exploited by introducing multivariable control of both bilayer thickness and penetration depth simultaneously, using a real-time computational model as an observer with adaptation informed by infrared measurements of external vial temperature. The proposed control scheme is tested on a laboratory low-energy ball milling system and demonstrated to effectively modulate power intensity and process duration to obtain the desired microstructure and material composition.
Interpreting chest X-rays via CNNs that exploit hierarchical disease dependencies and uncertainty labels
(2020) Hieu Pham; Tung Le; Dat Tran; Dat Ngo; Ha Nguyen
Chest radiography is one of the most common types of diagnostic radiology exams, which is critical for screening and diagnosis of many different thoracic diseases. Specialized algorithms have been developed to detect several specific pathologies such as lung nodules or lung cancer. However, accurately detecting the presence of multiple diseases from chest X-rays (CXRs) is still a challenging task. This paper presents a supervised multi-label classification framework based on deep convolutional neural networks (CNNs) for predicting the presence of 14 common thoracic diseases and observations. We tackle this problem by training state-of-the-art CNNs that exploit hierarchical dependencies among abnormality labels. We also propose to use the label smoothing technique for a better handling of uncertain samples, which occupy a significant portion of many CXR datasets. Our model is trained on over 200,000 CXRs of the recently released CheXpert dataset and achieves a mean area under the curve (AUC) of 0.940 in predicting 5 selected pathologies from the validation set. This is the highest AUC score yet reported to date. The proposed method is also evaluated on the independent test set of the CheXpert competition, which is composed of 500 CXR studies annotated by a panel of 5 experienced radiologists. The performance is on average better than 2.6 out of 3 other individual radiologists with a mean AUC of 0.930, which ranks first on the CheXpert leaderboard at the time of writing this paper.
Global Context Aware Convolutions for 3D Point Cloud Understanding
(2020) Zhiyuan Zhang; Son Hua; Wei Chen; Yibin Tian; Sai-Kit Yeung
Recent advances in deep learning for 3D point clouds have shown great promises in scene understanding tasks thanks to the introduction of convolution operators to consume 3D point clouds directly in a neural network. Point cloud data, however, could have arbitrary rotations, especially those acquired from 3D scanning. Recent works show that it is possible to design point cloud convolutions with rotation invariance property, but such methods generally do not perform as well as translation-invariant only convolution. We found that a key reason is that compared to point coordinates, rotation-invariant features consumed by point cloud convolution are not as distinctive. To address this problem, we propose a novel convolution operator that enhances feature distinction by integrating global context information from the input point cloud to the convolution. To this end, a globally weighted local reference frame is constructed in each point neighborhood in which the local point set is decomposed into bins. Anchor points are generated in each bin to represent global shape features. A convolution can then be performed to transform the points and anchor features into final rotation-invariant features. We conduct several experiments on point cloud classification, part segmentation, shape retrieval, and normals estimation to evaluate our convolution, which achieves state-of-the-art accuracy under challenging rotations.
Synthesis of bulk reactive Ni–Al composites using high pressure torsion
(2020-10-15) Renk, Oliver; Tkadletz, Michael; Kostoglou, Nikolaos; Gunduz, Emre Ibirahim; Fezzaa, Kamel; Sun, Tao; Stark, Andreas; Doumanidis, C. Charalabos; Eckert, Jurgen; Pippan, Reinhard; Mitterer, Christian; Rebholz, Claus
Self-propagating exothermic reactions, for instance, in the nickel aluminum (NieAl) system, have been widely studied to create high-performance intermetallic compounds or for in-situ welding. Their easy ignition once the phase spacing is reduced below the micron scale makes top-down methods like high-energy ball milling, ideal to fabricate such reactive nanostructures. A major drawback of ball milling is the need of a sintering step to form bulk pieces of the reactive material. However, this is not possible, as the targeted reactions would already proceed. Therefore, we investigate the ability of high-pressure torsion as an alternative process, capable to produce bulk nanocomposites from powder mixtures. Severe straining of powder mixtures with a composition of 50 wt% Ni and 50 wt% Al enables the fabrication of self-reactive bulk samples with microstructures similar to those obtained from ball milling or magnetron sputtering. Samples deformed at ambient temperature are highly reactive and can be ignited significantly below the Al melting point, finally predominantly consisting of Al3Ni2 and Al3Ni, independent of the applied strain. Although the reaction proceeds first at the edge of the disk, the strain gradient present in the disks does not prevent the reaction of the whole sample.
An experimental method to characterize the relationship between aperture image and ray directions in microscope optics
(2020-10-21) Mai Tran; Oldenbourg, Rudolf
We propose a direct experimental method to calibrate the relationship between ray directions in object space and their positions in the aperture plane of a light field microscope. The calibration improves the interpretation of light field images, which contain information from both types of image planes, the field plane and the aperture plane of the ray path in the microscope. Our method is based on the diffraction of line gratings of known periodicities and provides accurate results with subpixel resolution. The method can be custom-tailored to most any optical configuration, including standard light microscopy setups, whenever correct mapping between ray parameters in the object/image plane and the aperture plane is needed.
Toward forecasting future day air pollutant index in Malaysia
(2020-10-26) Wong, Kok-Seng; Chew, Yee Jian; Ooi, Shih Yin; Pang, Ying Han
The association of air pollution and the magnitude of adverse health effects are receiving close attention from the world. The effects of air pollution were found to be most significant for children, elderly, and patients with preexisting respiratory problems. The existing API forecast system is capable of predicting air quality based on pollutant concentrations before critical levels of air pollution are exceeded. However, there is no API forecasting system available in Malaysia that can predict the coming day API readings. This paper aims to propose an API forecast system that utilizes the hourly API in Malaysia to predict the next-day API. The proposed solution allows sensitive populations to plan ahead of their daily activities and provide governments with information for public health alerts. We also propose strategies for aggregated-level predictions within the region. Nevertheless, it can be extended across the region, especially in the less economically developed regions across the world. We conduct experiments on the public API dataset to demonstrate the viability of the proposed solution.

Browse

Browsing The College of Engineering and Computer Science by Issue Date

Results Per Page

Sort Options