Department of Computer Science & Engineering

&ldquo;IIT Jodhpur is committed to impart quality education to facilitate the creation of qualified, competent and responsible human resources to meet the emerging technological challenges of the world. It is working towards creating an equitable learning platform that is positive, accessible and effective in the face of the modern societal and scientific advancements.

Flexibility in curriculum, in all its academic programs, encompassing elements of teaching, research, creativity, innovation and entrepreneurship is the most unique feature of IIT Jodhpur. The Department of Science and Technology (DST), under the National Mission on Cyber Physical Systems, has identified IIT Jodhpur as a Technology &amp; Innovation Hub in the area of Augmented Reality and Virtual Reality. The institute has also been nominated as the coordinator for the Jodhpur City and Knowledge Cluster by the Office of Principal Scientific Adviser, Government of India.

The sprawling 852-acre IIT Jodhpur campus is a majestic tribute to the deep-rooted diverse cultural heritage of the region. The campus is an architectural masterpiece, which is slated to become an international exemplar of sustainability with net-zero energy, and advanced water and waste management strategies.&rdquo;

IIT Jodhpur

Objective: Effective connectivity (EC) is the methodology for determining functional-integration among the functionally active segregated regions of the brain. By definition [1] EC is 'the causal influence exerted by one neuronal group on another' which is constrained by anatomical connectivity (AC) (axonal connections). AC is necessary for EC but does not fully determine it, because synaptic communication occurs dynamically in a context-dependent fashion. Although there is a vast emerging evidence of structure-function relationship using multimodal imaging studies, till date only a few studies have done joint modeling of the two modalities: functional MRI (fMRI) and diffusion tensor imaging (DTI). We aim to propose a unified probabilistic framework that combines information from both sources to learn EC using dynamic Bayesian networks (DBNs). Method: DBNs are probabilistic graphical temporal models that learn EC in an exploratory fashion. Specifically, we propose a novel anatomically informed (AI) score that evaluates fitness of a given connectivity structure to both DTI and fMRI data simultaneously. The AI score is employed in structure learning of DBN given the data. Results: Experiments with synthetic-data demonstrate the face validity of structure learning with our AI score over anatomically uninformed counterpart. Moreover, real-data results are cross-validated by performing classification-experiments. Conclusion: EC inferred on real fMRI-DTI datasets is found to be consistent with previous literature and show promising results in light of the AC present as compared to other classically used techniques such as Granger-causality. Significance: Multimodal analyses provide a more reliable basis for differentiating brain under abnormal/diseased conditions than the single modality analysis. © 1964-2012 IEEE.

World University root node

School of Artificial Intelligence and Data Science (AIDE)

IEEE Transactions on Biomedical Engineering

Tractography-Based Score for Learning Effective Connectivity from Multimodal Imaging Data Using Dynamic Bayesian Networks

Advancements in smartphone applications have empowered even non-technical users to perform sophisticated operations such as morphing in faces as few tap operations. While such enablements have positive effects, as a negative side, now anyone can digitally attack face (biometric) recognition systems. For example, face swapping application of Snapchat can easily create 'swapped' identities and circumvent face recognition system. This research presents a novel database, termed as SWAPPED - Digital Attack Video Face Database, prepared using Snap chat's application which swaps/stitches two faces and creates videos. The database contains bonafide face videos and face swapped videos of multiple subjects. Baseline face recognition experiments using commercial system shows over 90% rank-1 accuracy when attack videos are used as probe. As a second contribution, this research also presents a novel Weighted Local Magnitude Pattern feature descriptor based presentation attack detection algorithm which outperforms several existing approaches. © 2017 IEEE.

IEEE International Joint Conference on Biometrics, IJCB 2017

SWAPPED! Digital face presentation attack detection via weighted local magnitude pattern

Forensic application of automatically matching skull with face images is an important research area linking biometrics with practical applications inforensics. It is an opportunity for biometrics and face recognition researchers to help the law enforcement and forensic experts in giving an identity to unidentified human skulls. It is an extremely challenging problem which is further exacerbated due to lack of any publicly available database related to this problem. This is the first research in this direction with a twofold contribution: (i) introducing the first of its kind skull-face image pair database, Identify Me, and (ii) presenting a preliminary approach using the proposed semi-supervised formulation of transform learning. The experimental results and comparison with existing algorithms showcase the challenging nature of the problem. We assert that the availability of the database will inspire researchers to build sophisticated skull-to-face matching algorithms. © 2017 IEEE.

On matching skulls to digital face images: A preliminary approach

Deep Learning in Biometrics

Deep Face-Representation Learning for Kinship Verification

Learning Representations for Unconstrained Fingerprint Recognition

Temporal segmentation of complex human action videos into action primitives plays a pivotal role in building models for human action understanding. Studies in the past have introduced unsupervised frameworks for deriving a known number of motion primitives from action videos. Our work focuses towards answering a question: Given a set of videos with humans performing an activity, can the action primitives be derived from them without specifying any prior knowledge about the count for the constituting sub-actions categories? To this end, we present a novel community detection-based human action segmentation algorithm. Our work marks the existence of community structures in human action videos where the consecutive frames around the key poses group together to form communities similar to social networks. We test our proposed technique over the stitched Weizmann dataset and MHADI01-s motion capture dataset and our technique outperforms the state-of-the-art techniques of complex action segmentation without the count of actions being pre-specified. © 2018 IEEE.

Proceedings - International Conference on Image Processing, ICIP

Unsupervised Temporal Segmentation of Human Action Using Community Detection

Self assemblies of (N2O4) salen-type bi-compartmental ligand (H2L°Me) [N,N′-Bis(3-methoxysalicylidenimino)-1,3-diaminopropane] afforded two series of complexes, a trinuclear nitrate [Cd3(L°Me)2(NO3)2] (1) and a dinuclear (η1-N3) [Zn2(L°Me)(η1-N3)2(CH3OH)2] (2). Elemental analysis, UV-visible and Fourier transform Infrared spectroscopic methods were employed successfully to characterize the complexes. Single-crystal X-ray studies reveal that complex 1 consists of planar trinuclear Cd(II)3 core coordinated by two fully deprotonated dianionic ligands [L1]2− and nitrato (NO3 −) bridging but complex 2 is a centrosymmetric hexacoordinated Zn(II) double end-on azide (η1-N3) with methanol as co-ligands. Both complexes satisfied a common 4-membered M(II)2(μ2-O)2 basic structural core. Fluorescence properties in DMSO solvent reveal that complexes enhance appreciably the fluorescence behavior over free salen-type ligand (H2L°Me). Complexes are optimized using Density functional theory calculated at B3LYP/TZVP level of theory to obtain insights into optimized structure, Frontier molecular orbital, Electrostatic potential maps, charge analyses gives an insight into the electronic structure and charge/potential distribution in the molecule etc. which substantiate the experimentally observed supramolecular interactions. Existing different non-covalent supramolecular interactions have been explained by means of Hirshfeld surface and 2D fingerprint plot analysis. Further theoretical explanations for the observed experimental electronic spectra of both complexes were explained by using Time-dependent density functional theory level of calculations. © 2018 Wiley-VCH Verlag GmbH &amp; Co. KGaA, Weinheim

Department of Chemistry

ChemistrySelect

Nitrato, Pseudohalo-Linked Zn(II)/Cd(II) Schiff-Base Complexes with 1,3-Diimine Spacer Group: Syntheses, Crystal Structures, DFT, TD-DFT and Fluorescence Studies

We are living in a world where the road infrastructure is something that can not be easily improved, that is why the field of Intelligent Transportation Systems (ITS) has arisen. Every year, the number of traffic accidents in the world is increasing. The biggest cause of the accident was due to human negligence. Vehicular Ad-hoc Networks (VANETs) is a growing interest and research area over recent years for that, it offers enhanced safety and non-safety applications for ITS. SD-VANETs is a technology that can provide tractability and programmability to networks. Moreover, we focused on recent proposals with the objective of enhancing SD-VANETs security and routing in a systematic and architectural way. Our scheme provides an identification mechanism for malicious vehicles using the trust based concept in a dynamic environment as well as reducing the overhead by avoiding those vehicles whose trust value is 0. For that, we presented two algorithms. Finally, we proved that finding malicious vehicles in SD-VANETs is an NP-Complete problem. Simulation results are examined to check, the usefulness of our new system. © 2018 IEEE.

International Conference on Information Networking

A trust based secure communication for software defined VANETs

The Road traffic modeling in Vehicular Ad Hoc Networks (VANETs) may require disseminating information to specific geographical areas like Road Intersection. The communications, network connectivity and routing are the central issues in VANETs. The Vehicular Ad Hoc Network (VANETs) have high mobility and high density of mobile nodes. Managing this mobility is very important to ensure routing efficiency and network connectivity in VANETs. VANETs are characterized by topology changes and link disconnections. The research challenges for applications of VANETs remains to design of intelligent algorithms for dynamic network connectivity and efficient routing problem in VANETs becomes a challenging event because of the active nature of mobile clients. In this paper, we have proposed an appropriate routing protocol for VANETs based on the dynamic network connectivity behavior of urban traffic at road crossings or Intersection. The dynamic network connectivity model for VANETs is based on parallel processing concept are essential for designing an efficient routing protocol for VANETs at Road Intersection in urban scenarios. Our simulation result showed that different control mechanisms are applied at road crossings and we likened the performance of our proposed scheme with the competitive schemes (Modeling Urban Traffic (MUT) Scheme, Dynamic Connectivity Scheme (DCS), Cellular Automaton Model for Car Traffic (CAMCT) and Dynamic Connectivity-Aware Routing for Internet-based Services (D-CAR)) in terms of Inter-vehicle spacing, number of vehicles in one road block and distance from road crossings or Intersection against varying node density. © 2018 Elsevier Ltd

Journal of Network and Computer Applications

Improvised dynamic network connectivity model for Vehicular Ad-Hoc Networks (VANETs)

d-Hitting Set and d-Set Cover are among the classical NP-hard problems. In this paper, we study variants of d-Hitting Set and d-Set Cover, which are called Partial d -Hitting Set (Partial d -HS) and Partial d -Exact Set Cover (Partial d -Exact SC), respectively. In Partial d -HS, given a universe U, a family F, of sets of size at most d over U, and integers k and t, the objective is to decide if there exists a (Formula Presented) of size at most k such that S intersects with at least t sets in F. We obtain a kernel for Partial d -HS in which the size of the universe is bounded by O(dt) and the size of the family is bounded by O(dt2). Using this result, we obtain a kernel for Partial Vertex Cover (PVC) with O(t) vertices, where t is the number of edges to be covered. Next, we study the Partial d -Exact SC problem, where, given a universe U, a family F, of sets of size exactly d over U, and integers k and t, the objective is to decide if there is (Formula Presented) of size at most k, such that S covers at least t elements in U. We design a kernel for Partial d -Exact SC in which sizes of the universe and the family are bounded by (Formula Presented). Finally, we study a special case of Partial d -HS, when d=2, and design an exact exponential time algorithm with running time (Formula Presented). © 2018, Springer International Publishing AG, part of Springer Nature.