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

Gender is one of the most common attributes used to describe an individual. It is used in multiple domains such as human computer interaction, marketing, security, and demographic reports. Research has been performed to automate the task of gender recognition in constrained environment using face images, however, limited attention has been given to gender classification in unconstrained scenarios. This work attempts to address the challenging problem of gender classification in multi-spectral low resolution face images. We propose a robust Class Representative Autoencoder model, termed as AutoGen for the same. The proposed model aims to minimize the intra-class variations while maximizing the inter-class variations for the learned feature representations. Results on visible as well as near infrared spectrum data for different resolutions and multiple databases depict the efficacy of the proposed model. Comparative results with existing approaches and two commercial off-the-shelf systems further motivate the use of class representative features for classification. © 2017 IEEE.

World University root node

Proceedings of the International Joint Conference on Neural Networks

Class representative autoencoder for low resolution multi-spectral gender classification

Recent interest and requirement of law enforcement agencies in matching composite sketches with digital images has instigated research in this important face recognition problem. In this paper, we propose feature extraction and matching algorithm using visual saliency and combination of texture features for matching composite sketches with digital photos. The attributes such as gender, ethnicity, and skin color are utilized for re-ordering the ranked list. Further, information from multiple experts such as multiple composite sketch generation tools or artists is combined for improving the matching performance. The results obtained on the extended PRIP database show that the proposed algorithm improves the state-of-art in matching composite sketch and digital face images and yields the rank 50 identification accuracy of 70.3% on a database of 1500 subjects. © 2016 Elsevier B.V. All rights reserved.

Information Fusion

Composite sketch recognition using saliency and attribute feedback

Restricted Boltzmann Machines (RBMs) have been extensively utilized in machine learning as core units in constructing deep learning architectures such as Deep Boltzmann Machines (DBMs) and Deep Belief Networks (DBNs). However, they are prone to overfitting and several regularization techniques have been proposed to mitigate this effect. In this paper, we propose the semi-supervised class sparsity signature based RBM formulation by combining unsupervised generative training of the RBM with a supervised sparsity regularizer. The proposed approach, termed as cssRBM, enforces sparsity at the class level to ensure that coherent and discriminative representations are learnt during training. Combining unsupervised learning with supervised learning allows the model to utilize external training data to learn better generative features while the supervised learning enables fine-tuning for discrimination using the learned features. We construct both DBMs and DBNs with cssRBM units and evaluate the performance on multiple publicly available benchmark datasets. Experiments on the MNIST and CIFAR-10 databases demonstrate that the proposed approaches are comparable with state-of-the-art deep learning architectures in the literature. We also evaluate the performance on one of the most challenging face databases, i.e., the Point and Shoot Challenge dataset. The results show that the proposed approaches improve state-of-the-art results by 15% on the PaSC database. © 2016 Elsevier Ltd

Pattern Recognition

Class sparsity signature based Restricted Boltzmann Machine

Building bilateral semantic associations between images and texts is among the fundamental problems in computer vision. In this paper, we study two complementary cross-modal prediction tasks: (i) predicting text(s) given a query image (“Im2Text”), and (ii) predicting image(s) given a piece of text (“Text2Im”). We make no assumption on the specific form of text; i.e., it could be either a set of labels, phrases, or even captions. We pose both these tasks in a retrieval framework. For Im2Text, given a query image, our goal is to retrieve a ranked list of semantically relevant texts from an independent text-corpus (i.e., texts with no corresponding images). Similarly, for Text2Im, given a query text, we aim to retrieve a ranked list of semantically relevant images from a collection of unannotated images (i.e., images without any associated textual meta-data). We propose a novel Structural SVM based unified framework for these two tasks, and show how it can be efficiently trained and tested. Using a variety of loss functions, extensive experiments are conducted on three popular datasets (two medium-scale datasets containing few thousands of samples, and one web-scale dataset containing one million samples). Experiments demonstrate that our framework gives promising results compared to competing baseline cross-modal search techniques, thus confirming its efficacy. © 2016

Computer Vision and Image Understanding

A support vector approach for cross-modal search of images and texts

Software Quality Attributes (QAs) can be categorised as either internal to the system as experienced by the developers or external to the system perceived by the end users. These QA categories have trade-off among them - an emphasis on internal QA may result in a compromise of an external QA. For example, there is a trade-off between maintainability and performance. Model-driven development approaches manage this trade-off and increase the degree of internal QA maintainability. In this work, we propose an ontology-based communication mechanism among software components to handle the trade-off. The approach increases the degree of internal QAs such as modifiability, maintainability, testability during the design and development phases without compromising the external QAs for the end users during the operation phase. We also evaluate a prototype system to validate the proposed approach using Software Architecture Analysis Method (SAAM). It is also easier to integrate into the software development life cycle as compared to existing model-driven approaches. © 2017 Association for Computing Machinery.

Proceedings of the 2017 Research in Adaptive and Convergent Systems, RACS 2017

Internal quality to external quality - An approach to manage conflicts

Multi-tenant architectural principle for cloud applications is a relatively new principle as compared to others such as multi- user and multi-instance. It allows to share a single instance of an application among multiple tenants with a high degree of customization. In this paper, we present a collection of patterns to design a multi-tenant application such that it can manage its tenants efficiently. The patterns described here are - multi-tenant, tenant manager, tenant observability, tenant separability and tenant elasticity. Each pattern covers the different aspects of systematic tenant management. © 2017 Copyright is held by the owner/author(s).

ACM International Conference Proceeding Series

Patterns for managing tenants in a multi-tenant application

Although visual object tracking algorithms are capable of handling various challenging scenarios individually, none of them are robust enough to handle all the challenges simultaneously. For any online tracking by detection method, the key issue lies in detecting the target over the whole frame and updating systematically a target model based on the last detected appearance to avoid the drift phenomenon. This paper aims at proposing a novel robust tracking algorithm by fusing the frame level detection strategy of tracking, learning, detection with the systematic model update strategy of Kernelized Correlation Filter tracker. The risk of drift is mitigated by the fact that the model updates are primarily driven by the detections that occur in the spatial neighborhood of the latest detections. The motivation behind the selection of trackers is their complementary nature in handling tracking challenges. The proposed algorithm efficiently combines the two state-of-the-art tracking algorithms based on conservative correspondence measure with strategic model updates, which takes advantages of both and outperforms them on their short ends by virtue of other. Extensive evaluation of the proposed method based on different metrics is carried out on the data sets ALOV300++, Visual Tracker Benchmark, and Visual Object Tracking. We demonstrated its performance in terms of robustness and success rate by comparing with state-of-the-art trackers. © 1992-2012 IEEE.

IEEE Transactions on Image Processing

Correlation-Based Tracker-Level Fusion for Robust Visual Tracking

We propose and demonstrate a compressive sensing (CS) framework for correlation holography. This is accomplished by adopting the principle of compressive sensing and thresholding in the two-point intensity correlation. The measurement matrix and the sensing matrix that are required for applying the CS framework here are systematically extracted from the random illuminations of the laser speckle data. Reconstruction results using CS, CS with thresholding, and intensity correlation are compared. Our study reveals that liminal CS requires far fewer samples for the reconstruction of the hologram and has wide application in image reconstruction. © 2017 Optical Society of America.

Applied Optics

Compressive correlation holography

Animal Biometrics

The Internet of Things(IoT) has been developing rapidly since the past decade. It is becoming a crucial part of everyday life. It is becoming a way of living in which machines perform activities on their own by communicating with each other, without the need for human beings to instruct them. In such a scenario, which concerns daily life, privacy, and secrecy should be of at most priority. Since in Internet of Things(IoT) several machines communicate with each other, mutual authentication and service discovery must be performed in such a way that each of the devices share information with only those devices that are genuine. Hence, private mutual authentication and automatic service discovery is important for the functioning of the IoT without human intervention. But, this comes at the cost of computation time. Thus, there is a balance to be found between the two. This can be achieved by introducing a trust attribute. We propose here that the distance between the client and server and the presence of a signed token with the client act as components of the trust attribute. Here we mention a hybrid approach that determines in real time which protocol to be used (SIGMA-I or policy hiding prefix based encryption) based on the proximity/distance between the entities involved and presence of a token with the client. © 2017 IEEE.