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

The present work is part of ongoing effort to improve the performance of Gujarati character recognition. In the recent advancement in kernel methods, the novel concept of multiple kernel learning(MKL) has given improved results for many problems. In this paper, we present novel application ofMKL for Gujarati character recognition. We have applied three different feature representations for symbols obtained after zone wise segmentation of Gujarati text. TheMKLbased classification is proposed, where the MKL is used for learning optimal combination of different features for classification. In addition MKL based classification results for different features is also presented. The multiclass classification is performed in Decision DAG framework. The comparison results in 1-Vs-1 framework and using KNN classifier is also presented. The experiments have shown substantial improvement in earlier results. © Copyright 2010 ACM.

World University root node

ACM International Conference Proceeding Series

Use of MKL as symbol classifier for Gujarati character recognition

Newborn swapping and abduction is a global problem and traditional approaches such as ID bracelets and footprinting do not provide the required level of security. This paper introduces the concept of using face recognition for identifying newborns and presents an automatic face recognition algorithm. The proposed multiresolution algorithm extracts Speeded up robust features and local binary patterns from different levels of Gaussian pyramid. The feature descriptors obtained at each Gaussian level are combined using weighted sum rule. On a newborn face database of 34 babies, the proposed algorithm yields rank-1 identification accuracy of 86.9%. © 2010 IEEE.

IEEE 4th International Conference on Biometrics: Theory, Applications and Systems, BTAS 2010

Face recognition for newborns: A preliminary study

The performance of a large scale biometric system may deteriorate over time as new individuals are continually enrolled. To maintain an acceptable level of performance, the classifier has to be re-trained offline in batch mode using both existing and new data. The process of re-training can be computationally expensive and time consuming. This paper presents a new biometric classifier update algorithm that incrementally re-trains the classifier using online learning and progressively establishes a decision hyperplane for improved classification. The proposed algorithm incorporates soft labels and granular computing in the formulation of a 2ν-Online Granular Soft Support Vector Machine (SVM) to re-train the classifier using only the new data. Granular computing makes it adaptive to local and global variations in data distribution, while soft labels provide resilience to noise. Each time data is acquired, new support vectors that are linearly independent are added and existing support vectors that do not improve the classifier performance are removed. This constrains the size of the support vectors and significantly reduces the training time without compromising the classification accuracy. The efficacy of the proposed online learning strategy is validated in a near infrared face verification application involving different covariates. The results obtained on a heterogeneous near infrared face database of 328 subjects show that in all experiments using different feature extraction and classification algorithms the proposed online 2ν-Granular Soft Support Vector Machine learning approach is 2-3 times faster while achieving a high level of accuracy similar to offline training using all data. © 2010 Elsevier B.V. All rights reserved.

Image and Vision Computing

Biometric classifier update using online learning: A case study in near infrared face verification

Face recognition systems has engrossed much attention and has been applied in various domains, primarily for surveillance, security, access control and law enforcement. In recent years much advancement have been made in face recognition techniques to cater to the challenges such as pose, expression, illumination, aging and disguise. However, due to advances in technology, there are new emerging challenges for which the performance of face recognition systems degrades and plastic/cosmetic surgery is one of them. In this paper we comment on the effect of plastic surgery on face recognition algorithms and various social, ethical and engineering challenges associated with it. © Springer-Verlag Berlin Heidelberg 2010.

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

Face recognition and plastic surgery: Social, ethical and engineering challenges

Many RFID protocols use cryptographic hash functions for their security. The resource constrained nature of RFID systems forces the use of light weight cryptographic algorithms. Tav-128 is one such 128-bit light weight hash function proposed by Peris-Lopez et al. for a low-cost RFID tag authentication protocol. Apart from some statistical tests for randomness by the designers themselves, Tav-128 has not undergone any other thorough security analysis. Based on these tests, the designers claimed that Tav-128 does not posses any trivial weaknesses. In this article, we carry out the first third party security analysis of Tav-128 and show that this hash function is neither collision resistant nor second preimage resistant. Firstly, we show a practical collision attack on Tav-128 having a complexity of 237 calls to the compression function and produce message pairs of arbitrary length which produce the same hash value under this hash function. We then show a second preimage attack on Tav-128 which succeeds with a complexity of 262 calls to the compression function. Finally, we study the constituent functions of Tav-128 and show that the concatenation of nonlinear functions A and B produces a 64-bit permutation from 32-bit messages. This could be a useful light weight primitive for future RFID protocols. © 2010 Springer-Verlag Berlin Heidelberg.

Cryptanalysis of Tav-128 hash function

The thinning methodology is novel in terms of its ability to incorporate character shape specific knowledge while constructing the thinned skeleton. But removal of spurious strokes or shape deformation in thinning is a difficult problem. In this paper, we have proposed a novel medial-axis based thinning strategy used for performing skeletonization of noisy character images. The proposed algorithm produces segmented strokes in vector form as a by-product. Hence further stroke segmentation is not required. Experiment is done on printed English, Bengali, Hindi, and Tamil characters and we obtain less spurious branches compared to other thinning methods without any post processing. We have concluded with a proposed methodology to extract structural features from thinned character images. This feature set improves the performance of existing OCR for Indian languages. © 2010 IEEE.

Proceedings - International Conference on Image Processing, ICIP

A medial axis based thinning strategy and structural feature extraction of character images

The SHREC'10 correspondence finding benchmark simulates a one-to-one shape matching scenario, in which one of the shapes undergoes multiple modifications and transformations. The benchmark allows evaluating how correspondence algorithms cope with certain classes of transformations and what is the strength of the transformations that can be dealt with. The present paper is a report of the 3D Shape Retrieval Contest 2010 (SHREC'10) correspondence finding benchmark results. ©The Eurographics Association 2010.

Eurographics Workshop on 3D Object Retrieval, EG 3DOR

SHREC'10 track: Correspondence finding

The paper presents three novel features for handwritten data based identity recognition. A novel framework for combining the features for identification is presented. The framework combines the features in kernel space in MKL based framework. The application of features individually and in combination is presented for writer recognition and signature verification. The writer recognition results have been presented for Devanagari script input and signature verification results have been presented for open dataset [1]. The experiments have shown encouraging results. © 2010 IEEE.

Proceedings - 12th International Conference on Frontiers in Handwriting Recognition, ICFHR 2010

Identity determination with offline handwritten input using multi kernel feature combination

In this paper, we present a technique for distributed selfcalibration of pan-tilt camera network using multi-layered belief propagation. Our goal is to obtain globally consistent estimates of the camera parameters for each camera with respect to a global world coordinate system. The network configuration changes with time as the cameras can pan and tilt. We also give a distributed algorithm for automatically finding which cameras have overlapping views at a certain point in time. We argue that using belief propagation it is sufficient to have correspondences between three cameras at a time for calibrating a larger set of (static) cameras with overlapping views. Our method gives an accurate and globally consistent estimate of the camera parameters of each camera in the network. © 2010 IEEE.

2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Workshops, CVPRW 2010

Distributed calibration of pan-tilt camera network using multi-layered belief propagation

Biometric fusion consolidates the output of multiple biometric classifiers to render a decision about the identity of an individual. We consider the problem of designing a fusion scheme when 1) the number of training samples is limited, thereby affecting the use of a purely density-based scheme and the likelihood ratio test statistic; 2) the output of multiple matchers yields conflicting results; and 3) the use of a single fusion rule may not be practical due to the diversity of scenarios encountered in the probe dataset. To address these issues, a dynamic reconciliation scheme for fusion rule selection is proposed. In this regard, the contribution of this paper is two-fold: 1) the design of a sequential fusion technique that uses the likelihood ratio test-statistic in conjunction with a support vector machine classifier to account for errors in the former; and 2) the design of a dynamic selection algorithm that unifies the constituent classifiers and fusion schemes in order to optimize both verification accuracy and computational cost. The case study in multiclassifier face recognition suggests that the proposed algorithm can address the issues listed above. Indeed, it is observed that the proposed method performs well even in the presence of confounding covariate factors thereby indicating its potential for large-scale face recognition. © 2010 IEEE.