Computer Science and Engineering

<p>Shiv Nadar University is a comprehensive, multidisciplinary, research-focused, and student-centric University offering a full range of academic programs at the undergraduate, postgraduate and doctoral level. With state-of-the-art infrastructure, the University comprises of academic wings, sophisticated labs, international standard sports facilities, amphitheaters, auditorium, conference rooms and smart classrooms. The University&rsquo;s goal is to become internationally recognized for the quality of its research and creative endeavors and their applicability to improving quality of life, generating new insights and expanding the boundaries of human knowledge creativity. Committed to excellence in teaching, research and service, the University aims to serve the higher education needs of India and the world beyond.</p>

Shiv Nadar University

Topic lifecycle analysis on Twitter, a branch of study that investigates Twitter topics from their birth through lifecycle to death, has gained immense mainstream research popularity. In the literature, topics are often treated as one of (a) hashtags (independent from other hashtags), (b) a burst of keywords in a short time span or (c) a latent concept space captured by advanced text analysis methodologies, such as Latent Dirichlet Allocation (LDA). The first two approaches are not capable of recognizing topics where different users use different hashtags to express the same concept (semantically related), while the third approach misses out the user’s explicit intent expressed via hashtags. In our work, we use a word embedding based approach to cluster different hashtags together, and the temporal concurrency of the hashtag usages, thus forming topics (a semantically and temporally related group of hashtags). We present a novel analysis of topic lifecycles with respect to communities. We characterize the participation of social communities in the topic clusters, and analyze the lifecycle of topic clusters with respect to such participation. We derive first-of-its-kind novel insights with respect to the complex evolution of topics over communities and time: temporal morphing of topics over hashtags within communities, how the hashtags die in some communities but morph into some other hashtags in some other communities (that, it is a community-level phenomenon), and how specific communities adopt to specific hashtags. Our work is fundamental in the space of topic lifecycle modeling and understanding in communities: it redefines our understanding of topic lifecycles and shows that the social boundaries of topic lifecycles are deeply ingrained with community behavior. © Springer International Publishing AG, part of Springer Nature 2018.

World University root node

(SoE) School of Engineering

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Topic lifecycle on social networks: Analyzing the effects of semantic continuity and social communities

With an alarming increase in obesity levels, even among children and adolescents, there is a need to take steps for increasing their motivation and overall physical activity levels. This paper proposes a way for concerned authorities in a school, to easily keep track of each student's physical activity during school hours and give feedback accordingly. The system uses a portable pedometer that includes a MPU6050 sensor module that uses an accelerometer and gyroscope to collect movement data from each student and stores it, via Wi-Fi communication on to the Thingspeak server. The system implements a k-NN algorithm that classifies the data for each student into activities of sitting, walking, running and cycling. The results of the analysis and feedback for each student are accessible through a web portal, hosted locally on a server located in the school premises. © 2018 IEEE.

INDICON 2018 - 15th IEEE India Council International Conference

Actiotracker: A Smart Pedometer for Monitoring and Analyzing Physical Activities of School Children

Estimating scene flow in RGB-D videos is attracting much interest of the computer vision researchers, due to its potential applications in robotics. The state-of-the-art techniques for scene flow estimation typically rely on the knowledge of scene structure of the frame and the correspondence between frames. However, with the availability of large RGB-D data captured from depth sensors, learning representations for estimation of scene flow has become possible. This paper introduces a first effort to apply a deep learning method for direct estimation of scene flow by presenting a fully convolutional neural network with an encoder-decoder (ED) architecture. The proposed network SceneEDNet involves estimation of three dimensional motion vectors of all the scene points from sequence of stereo images. The training for direct estimation of scene flow is done using consecutive pairs of stereo images and corresponding scene flow ground truth. © 2018 IEEE.

2018 15th International Conference on Control, Automation, Robotics and Vision, ICARCV 2018

SceneEDNet: A Deep Learning Approach for Scene Flow Estimation

It is generally believed that the existence of all organisms present on this earth has their point of convergence in a common gene pool. The current species passed through an evolutionary process which is still underway. The theoretical assumptions relating to the common descent of all organisms are based on four simple facts: first, they had wide geographical dispersal; second, the different life forms were not remarkably unique and did not possess mutually exclusive characteristics; third, some of their attributes which apparently served no purpose had an uncanny similarity with some of their lost functional traits; and last, based on their common attributes these organisms can be put together into a well-defined, hierarchical and coherent group, like a family tree. Phylogenetic networks are the main tools that can be used to represent biological relationship between different species. Biologists, mathematicians, statisticians, computer scientists and others have designed various models for the reconstruction of evolutionary networks and developed numerous algorithms for efficient predictions and analysis. Even though these problems have been studied for a very long time, but the computational model built to solve the biological problems fail to give accurate results while working on real biological data, which could be due to the premises on which the model is based. The objective of this paper is to test and analyse the transmission of commonly occurring diseases to fit into more realistic models. The problems are not only important because we need to know how they came into existence and how they migrated, but also helpful for the treatment of such diseases and drug discovery. © 2018, Springer Nature Singapore Pte Ltd.

Advances in Intelligent Systems and Computing

Analysing the genetic diversity of commonly occurring diseases

A theoretical model based on the generalized neuronal model with distributed delay (GNMDD) is proposed to generate multimodal/bimodal inter-spike interval (ISI) distribution. The distributed delay is assumed to be general in the sense that it is a linear combination of gamma distributed weak and strong memory kernels. It is found that the expected membrane potential in the subthreshold regime exhibits damped oscillatory behavior. This causes the ISI pattern to become bimodal/multimodal. Furthermore, the effect of external damped oscillatory current in the GNMDD model is investigated. Based on extensive simulation study of a set of coupled stochastic differential equations in the presence of a threshold, a vast amount of spiking data are generated. Using the non-parametric modeling approach, the optimal bin width of the histogram is obtained by using the Freedman-Diaconis rule. It is to be noted that, with the increasing amplitude of the damped oscillatory current, the multimodal ISI distribution changes to unimodal ISI distribution when the magnitude of external current reaches a critical value. It is found that the entropy also shows a sudden transition around the critical point. This phenomenon is akin to a phase transition. © 2018 IEEE.

IEEE Transactions on Nanobioscience

Suppression of multimodality in Inter-Spike interval distribution: Role of external damped oscillatory input

A new approximate expression for the probability density function (pdf) of K-distribution is proposed. The approximate pdf is found to be in good agreement with the exact analytical closed-form expression over the desired range of scintillation index lying between 2 and 3. Employing the proposed pdf, the expression in respect of two of the QoS measures of fade probability and bit error rate (BER) valid under various modulation schemes is given. Using the generalised expression of BER, plots under ON-OFF keying and binary phase shift keying modulation schemes are shown. The resulting expressions provide a better understanding in contrast to the exact results obtained in terms of Meijer G functions. Numerical computations are carried out to demonstrate the efficacy of approximate results. © The Institution of Engineering and Technology 2018.

Physics

(SoNS) School of Natural Sciences

IET Communications

New approximation for pdf of K- distribution: Analytical study of QoS parameters in free space optical communication

The presence of haze within the atmospheric medium degrades the quality of videos captured by camera sensors. The expulsion of haze, referred to as dehazing, is typically performed subject to the physical degradation display, that involves an explanation of an ill-posed inverse drawback. A few efforts have been made for image dehazing, whereas, video dehazing still remains an unexplored area of research. This paper proposes an approach for video dehazing combining the concepts of single image dehazing, optical stream estimation and Markov Random Field (MRF). The proposed method enhances the temporal and spatial coherence of the hazy video. Assuming that the dark channel of the haze-free picture is zero, we acquire the raw transmission map. In the proposed approach, we focus on the raw transmission map obtained from the dark channel prior using guided filter. We assess the forward and reverse optical streams between the neighboring frames to locate individual pixels using Linear Discriminant Analysis. The color of the haze-free pixels in the frames is approximated by a few hundred discrete colors, which generate a fixed cluster in space and the directions of the pixel. The pixels at a given cluster are spread and can be determined by analyzing the forward and in reverse optical frames to predict its value after haze removal. Largest Margin Nearest Neighbor (LMNN) algorithm is applied to get the smooth transmission map of the foggy frames of the video to approximate the pixel value in the RGB space. The stream fields are utilized in an augmented MRF model on the transmission guide obtained to enhance the temporal and the spatial coherence of the transmission. The proposed method is compared against the state-of-the-art on both real and synthetic videos to preserve the information optimally. © 2018 ACM.

ACM International Conference Proceeding Series

Video Dehazing using LMNN with respect to Augmented MRF

In this paper we present a novel methodology for recognizing human activity in Egocentric video based on the Bag of Visual Features. The proposed technique is based on the assumption that, only a portion of the whole video can be sufficient to identify an activity. Rather, we argue that, for activity recognition in egocentric videos, the proposed approach performs better than any deep learning based method. Because, in egocentric videos, often the person wiring the sensor, becomes static for long time, or moves his head frequently. In both the cases, it becomes difficult to learn the spatiotemporal pattern of the video during action. The proposed approach divides the video into smaller video segments called Video Units. Spatio-temporal features extracted from the units, are clustered to construct the dictionary of Action Units (AU). The AUs are ranked based upon their score of likeliness. The scores are obtained by constructing a weighted graph with the AUs as vertices and edge weights calculated based on the frequencies of occurrences of the AUs during the activity. The less significant AUs are pruned out from the dictionary, and the revised dictionary of key AUs are used for activity classification. We test our approach on benchmark egocentric dataset and achieve a good accuracy. © 2018 ACM.

Activity Recognition in Egocentric Videos Using Bag of Key Action Units

In this paper we present a novel analytical expression for the bit error rate under ON-OFF Keying (OOK) modulation for Free Space Optical Communication using α-μ distribution. We perform numerical comparison between the PDF, CDF and BER using α-μ and Gamma-Gamma distribution. The relative error between PDF and CDF of Gamma Gamma and α-μ demonstrates the viability of α-μ as a alternative to Gamma-Gamma distribution. © 2018 IEEE.

2018 27th Wireless and Optical Communication Conference, WOCC 2018

Exact analytical BER expression in FSO using α-μ Distribution proposed as an approximation of gamma-gamma PDF

A theoretical investigation is presented that characterizes the emerging sub-threshold membrane potential and inter-spike interval (ISI) distributions of an ensemble of IF neurons that group together and fire together. The squared-noise intensity σ2 of the ensemble of neurons is treated as a random variable to account for the electrophysiological variations across population of nearly identical neurons. Employing superstatistical framework, both ISI distribution and sub-threshold membrane potential distribution of neuronal ensemble are obtained in terms of generalized K-distribution. The resulting distributions exhibit asymptotic behavior akin to stretched exponential family. Extensive simulations of the underlying SDE with random σ2 are carried out. The results are found to be in excellent agreement with the analytical results. The analysis has been extended to cover the case corresponding to independent random fluctuations in drift in addition to random squared-noise intensity. The novelty of the proposed analytical investigation for the ensemble of IF neurons is that it yields closed form expressions of probability distributions in terms of generalized K-distribution. Based on a record of spiking activity of thousands of neurons, the findings of the proposed model are validated. The squared-noise intensity σ2 of identified neurons from the data is found to follow gamma distribution. The proposed generalized K-distribution is found to be in excellent agreement with that of empirically obtained ISI distribution of neuronal ensemble. © 2018 Elsevier B.V.