(SoE) School of Engineering

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.

Shiv Nadar University

The present study aims at conducting a comprehensive seismic risk assessment for the North Eastern Region of India at regional and sub-regional levels by integrating probabilistic seismic hazard and social vulnerability assessments. Bedrock-level peak ground acceleration varied from 0.14 to 0.69g for the return period of 475 years. Using PCA, the social vulnerability index (SVI) was generated considering district-level socioeconomic indicators. Built environment quality, illiteracy, access to amenities, dependent population, and employment opportunities contributed to high SVI. Most vulnerable districts were concentrated in the Brahmaputra floodplains, Tripura fold belt, and Imphal valley. At the regional level, significant parts of Assam, Meghalaya, Arunachal Pradesh, and Tripura lie in moderate to very high-risk zones. At the sub-regional level, Nagaland accounts for the highest proportion of areas in high to very high-risk zones. The findings will aid site-specific resilient infrastructure design, disaster risk reduction, and effective resource allocation for the risk-prone areas. © 2022 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.

World University root node

Civil Engineering

Sustainable and Resilient Infrastructure

Seismic risk assessment for the North Eastern Region of India by integrating seismic hazard and social vulnerability

The efficacy of Action Recognition methods depends upon relevance of the action (Verb) with respect to the subject (Noun). Existing methods overlook the suitability of Noun-Verb combination in defining an action. In this work, we propose an algorithm called Reduced Verb Set Generator (RVSGen) to reduce the number of possible verbs related to the actions, based upon the relevance of noun-verb combination. A dual modal fusion model for egocentric activity recognition is proposed here to combine the features extracted from the RGB channels and the Optical flow vectors for an egocentric video to recognize the human activity. Unlike state-of-the-art methods, where the key objects and temporal cues are extracted simultaneously, the proposed model first extracts the spatial features i.e., object information (Noun) from the RGB channels and then relates the Noun with the suitable action (Verb) obtained from motion information (Optical Flow). The verbs are predicted by an ConvLSTM architecture with the help of a modified softmax. The notion behind the modified softmax is to estimate the probability distribution with the reduced verb set obtained from RVSGen and the feature vector obtained from the ConvLSTM. With the help of an end-to-end trained architecture, the noun and verb are predicted which are then concatenated together constituting an action. The experiments are performed on a benchmark dataset. The results show the efficacy of the proposed method compared to the state-of-the-art. The codes related to this work can be found at: \url{https://github.com/mpLogics/EgoAR-RVSGen}.

Computer Science and Engineering

IJCNN

First-person Activity Recognition by Modelling Subject-Action Relevance

Recent years have been witnessing significant developments in research in human action recognition based on skeleton data. The graphical representation of the human skeleton, available with the dataset, provides opportunity to apply Graph Convolutional Networks (GCN), to avail efficient analysis of deep spatial-temporal information from the joint and skeleton structure. Most of the current works in skeleton action recognition use the temporal aspect of the video in short-term sequences, ignoring the long-term information present in the evolving skeleton sequence. The proposed long-term Spatio-temporal Contrastive Learning framework for Skeleton Action Recognition uses an encoder-decoder module. The encoder collects deep global-level (long-term) information from the complete action sequence using efficient self-supervision. The proposed encoder combines knowledge from the temporal domain with high-level information of the relative joint and structure movements of the skeleton. The decoder serves two purposes: predicting the human activity and predicting skeleton structure in the future frames. The decoder primarily uses the high-level encodings from the encoder to anticipate the action. For predicting skeleton structure, we extract an even deeper correlation in the Spatio-temporal domain and merge it with the original frame of the video. We apply a contrastive framework in the frame prediction part so that similar actions have similar predicted skeleton structure. The use of the contrastive framework throughout the proposed model helps achieve exemplary performance while employing a self-supervised aspect to the model. We test our model on the NTU-RGB-D-60 dataset and achieve state-of-the-art performance. The codes related to this work are available at: \url{https://github.com/AnshulRustogi/Long-Term-Spatio-Temporal-Framework}.

Long-term Spatio-temporal Contrastive Learning framework for Skeleton Action Recognition

Single image dehazing is a popular research topic among the researchers in computer vision, machine learning, image processing, and graphics. Most of the recent methods for single image dehazing are based upon supervised learning set up. However, supervised methods require annotation of the data, which often makes the dehazing methods biased towards the manual annotation errors. Unsupervised methods are more likely to produce realistic, clear images. However, fewer efforts are found in the literature for single image dehazing in unsupervised set up. We propose an enhanced CycleGAN architecture for Unpaired single image dehazing, with an attention-based transformer architecture embedded in the generator. The proposed transformer comprises three components: 1) A Feature Attention (FA) block combining channel attention and pixel attention mechanism, 2) A Dynamic feature enhancement block for dynamically capturing the spatial structured features and 3) An adaptive mix-up module to preserve the flow of shallow features from downsampling. Experiments on the benchmark datasets show the efficacy of the proposed method. Codes for this work are available in the link: \url{https://github.com/rsjai47/Attention-Based-CycleDehaze}.

Attention-based Single Image Dehazing Using Improved CycleGAN

People often query various Points of Interest (POIs) to plan their city tour itineraries using location-based services that provide reviews and ratings of various attributes of a POI. Most traditional location-based recommendation systems focus on various attributes like rating, cost, and social similarities of either POIs, users, or both but do not focus on efficient retrieval of POIs. This paper proposes a spatial index structure uR-tree (utility-based R-tree) for efficiently processing a location-based query implied in a location-aware recommender system to select multiple POIs which a user can visit in a tour. The uR-tree is constructed considering a change in static attributes like availability of parking, basic amenities, etc., and dynamic attributes like ticket price, congestion, etc., of each POI with the aim to maximize a user’s experience. At first, the utility of each POI is calculated at different points of time based on its static and dynamic attributes. A uR-tree is constructed for the systematic indexing of the utility of each POI, and an algorithm is applied for the retrieval of top-k maximum utility POI in different time-intervals. Then, we propose two Utility-based trip scheduling algorithms for distance-based travel recommendations. The experiments were conducted on real-world location-based social network data sets and show that the proposed scheme has a lower response time, Disk I/Os, and a higher query success rate as compared to state-of-the-art, R-tree index structure.

Multimedia tools and Application

uR-tree: a spatial index structure for handling multiple point selection queries

Fantasy Sports are becoming more and more popular these days, hence the race to crack it is more trending than ever. In this paper, we focus on cricket (IPL) and Dream11. Using advanced statistical and graphical models, and new performance metrics for batting and bowling we aim to build a model that can predict the top performing 11 players out of the two teams. This involves predicting the player performance and selecting the best 11 while complying with league constraints. The proposed model on an average predicts 70% of the players from the Dream Team.

Research Challenges in Information Science

Team Selection Using Statistical and Graphical Approaches for Cricket Fantasy Leagues

We propose a novel video sampling scheme for human action recognition in videos, using Gaussian Weighing Function. Traditionally in deep learning-based human activity recognition approaches, either a few random frames or every $k^{th}$ frame of the video is considered for training the 3D CNN, where $k$ is a small positive integer, like $4$, $5$, or $6$. This kind of sampling reduces the volume of the input data, which speeds-up the training network and also avoids overfitting to some extent, thus enhancing the performance of the 3D CNN model. In the proposed video sampling technique, consecutive $k$ frames of a video are aggregated into a single frame by computing a Gaussian-weighted summation of the $k$ frames. The resulting frame preserves the information in a better way than the conventional approaches and experimentally shown to perform better. In this paper, a 3-Dimensional deep CNN is proposed to extract the spatio-temporal features and follows Long Short-Term Memory (LSTM) to recognize human actions. The proposed 3D CNN architecture is capable of handling the videos where the camera is placed at a distance from the performer. Experiments are performed with KTH, WEIZMANN, and CASIA-B Human Activity and Gait datasets, whereby it is shown to outperform state-of-the-art deep learning based techniques. We achieve 95.78%, 95.27%, and 95.27% over the KTH, WEIZMANN, and CASIA-B human action and gait recognition datasets, respectively. &nbsp;&nbsp;

Multimedia Tools and Applications

An Information-rich Sampling Technique over Spatio-Temporal CNN for Classification of Human Actions in Videos

Mechanical Engineering

Environmental Science and Pollution Research

Review on performance assessment of solar stills using computational fluid dynamics (CFD)

Fulltext

Scientific Reports

Author Correction: Novel hybrid method to additively manufacture denser graphite structures using Binder Jetting (Scientific Reports, (2021), 11, 1, (2438), 10.1038/s41598-021-81861-w)

Cognitive activity prediction (CAP) from electroencephalogram (EEG) signals is progressively utilized in the field of brain-computer interface (BCI) and mental health management. Various machine and deep learning methods have been proposed recently for CAP. However, since Internet-of-Things-based real-time BCI systems demand low latency, power, and portability, these methods need to be deployable on resource-constrained edge devices. Towards this aspect, we propose a real-time implementation of a lightweight 1-D convolutional neural network on an Arduino Due microcontroller for CAP from EEG signals. The performance evaluation on two public datasets and one real-time recorded dataset indicates that the proposed work achieves subject-independent prediction accuracies of 99.30%, 82.50%, and 99.02% in these datasets. Furthermore, the prediction of real-time recorded EEG signals is accurate for majority of the subjects. The proposed work outperforms the existing techniques and achieves low power consumption of $\text{0.63}\,\text{W}$ in real-time on-device implementation with an average latency of $\text{455.12}\,\text{ms}$ in model deployment, test output prediction, and activity-based transmission. © 2017 IEEE.