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 Opportunistic Routing (OR) in a dynamic vehicular network increases the wireless network performance. It advantages from the broadcast nature of vehicular nodes to progress performance of the network. The primary responsibility of Opportunistic Routing is its capability to hear the communicated message and to manage among relaying vehicular nodes. In this work, we suggest a novel OR technique for Vehicular Ad-Hoc Networks (VANETs) to overcome the routing problem in VANETs. In this work, we concentrate on our consideration around the incorporation of Vehicular networks and Opportunistic Routing, which is the concept that we call as Vehicular Opportunistic Routing. To deal with the dynamic nature of vehicular node, we have recommended a innovative performance measure which reflects the location of the dynamic vehicular nodes, the link delivery probability(LDP) between the dynamic vehicular nodes, the number of the neighboring vehicular nodes in the highway/intersection, and the predicted current position of the highly dynamic vehicular nodes. We compared our proposed scheme with existing Opportunistic Routing schemes, and verified that our scheme performs better in message/packet delivery ratio (MDR/PDR), end-to-end(E-2-E) latency, and expected number of transmissions(ETX) from the source vehicular node to the destination vehicular node in terms of the Manhattan Grid Mobility Model. In our proposed approach, we have used Dijkstra Algorithm with binary heap for candidate selection. Using this approach, the time complexity of candidate selection is more efficient compared to existing schemes. © 2019 IEEE.

2019 2nd International Conference on Advanced Computational and Communication Paradigms, ICACCP 2019

An efficient opportunistic routing algorithm in vehicular adhoc networks(VANETs)

We construct a family of planar graphs {G-n}-n≥4, where G-n has n vertices including a source vertex s, a sink vertex t, and edge weights that change linearly with a parameter λ such that, as λ varies in (-∞,+∞), the piece-wise linear cost of the shortest path from s to t has nΩ(logn) pieces. This shows that lower bounds obtained by Carstensen (1983) and Mulmuley & Shah (2001) for general graphs also hold for planar graphs, refuting a conjecture of Nikolova (2009). Gusfield (1980) and Dean (2009) showed that the number of pieces for every n-vertex graph with linear edge weights is nlog n+O(1). We generalize this result in two ways. (i) If the edge weights vary as a polynomial of degree at most d, then the number of pieces is nlog n+(α(n)+O(1))d, where α(n) is the inverse Ackermann function. (ii) If the edge weights are linear forms of three parameters, then the number of pieces, appropriately defined for R3, is n (log n)2+O(log n). © 2019 IEEE.

World University root node

Proceedings - Annual IEEE Symposium on Foundations of Computer Science, FOCS

Parametric shortest paths in planar graphs

In this paper, a large scale public dataset containing floor plan images and their annotations is presented. BRIDGE (Building plan Repository for Image Description Generation, and Evaluation) dataset contains more than 13000 images of the floor plan and annotations collected from various websites, as well as publicly available floor plan images in the research domain. The images in BRIDGE also has annotations for symbols, region graphs, and paragraph descriptions. The BRIDGE dataset will be useful for symbol spotting, caption and description generation, scene graph synthesis, retrieval and many other tasks involving building plan parsing. In this paper, we also present an extensive experimental study for tasks like furniture localization in a floor plan, caption and description generation, on the proposed dataset showing the utility of BRIDGE. © 2019 IEEE.

Department of Mathematics

Proceedings of the International Conference on Document Analysis and Recognition, ICDAR

BRIDGE: Building plan repository for image description generation, and evaluation

In the recent past, due to the boost in digitization a huge volume of data (BigData) related to building and architectures has been accumulated. The sources of such data are interviews, blogs, websites, etc. These descriptions has details about the interior of a building. For humans it is easy to interpret and imagine its structure and arrangements of furniture. Automatic synthesis of real-world images from text descriptions has been explored in the computer vision community. However, there is no such attempt in the area of document images, like floor plans. Even though, Floor plan synthesis from sketches, as well as data-driven models, were proposed earlier, this is the first attempt to automatically render building floor plan images from textual description. Here, the input is a natural language description of the internal structure and furniture arrangements within a house, and the output is the 2D floor plan image of the same. We have experimented on publicly available benchmark floor plan datasets. We were able to render realistic synthesized floor plan images from the description written in English. © 2019 IEEE.

Proceedings - 2019 IEEE 5th International Conference on Multimedia Big Data, BigMM 2019

A framework for the conversion of textual bigdata into 2d architectural floor plan

We show that given an embedding of an O(log n) genus graph G and two vertices s and t in G, deciding if there is a path from s to t in G is in unambiguous logarithmic space. Unambiguous computation is a restriction of nondeterministic computation where the nondeterministic machine has at most one accepting computation path on each input. An important fundamental question in computational complexity theory is whether this is an actual restriction or are unambiguous computations as powerful as general nondeterminism. We investigate this problem in the domain of logarithmic space bounded computations, where the corresponding unambiguous and general nondeterministic classes are UL and NL respectively. In 1997 Reinhardt and Allender showed that NL and UL are equal in a non-uniform model. More specifically they showed that if one can efficiently construct an O(log n)-bit min-unique weight function for a graph, then these classes are equal unconditionally as well. In other words, they gave a UL algorithm to solve reachability in graphs with a min-unique weight assignment. Using this approach reachability in various classes of graphs such as planar graphs, constant genus graphs, minor free graphs, etc., have been shown to be in UL by devising min-unique weight functions for those classes. In this paper we improve these results by constructing a min-unique weight function for O(log n) genus graphs. We define signature of a path in a graph as the parity of the number of crossings of that path with respect to each handle of the surface on which the graph is embedded. We construct our weight function in two steps. First we ensure that between any pair of vertices, amongst all paths having the same signature, the minimum weight path is unique. Now since in a genus g graph there are 2^{2g} many possible signatures, we use the hashing scheme of Fredman, Komlós and Szemerédi to isolate a unique minimum weight path among these 2^{2g} many paths isolated in the first step.

36th International Symposium on Theoretical Aspects of Computer Science …

Reachability in O (log n) Genus Graphs is in Unambiguous Logspace

The catalytic Turing machine is a model of computation defined by Buhrman, Cleve, Koucký, Loff, and Speelman (STOC 2014). Compared to the classical space-bounded Turing machine, this model has an extra space which is filled with arbitrary content in addition to the clean space. In such a model we study if this additional filled space can be used to increase the power of computation or not, with the condition that the initial content of this extra filled space must be restored at the end of the computation. In this paper, we define the notion of unambiguous catalytic Turing machine and prove that under a standard derandomization assumption, the class of problems solved by an unambiguous catalytic Turing machine is same as the class of problems solved by a general nondeterministic catalytic Turing machine in the logspace setting.

39th IARCS Annual Conference on Foundations of Software Technology and …

Unambiguous Catalytic Computation

Gene Ontology (GO) is a taxonomy of biological terms related to the properties of genes and gene products. It can be used to define a similarity measure between two gene products and assign a confidence score to protein-protein interactions (PPIs). GO is being evolved regularly by the addition/deletion/merging of terms. However, there is no study which evaluates the robustness of a particular similarity measure over the evolution of GO. By robustness of a similarity measure, we mean it should either improve or keep its performance similar over the evolution of GO. In this paper, we systematically study the same for the task of scoring confidence of PPIs using GO-based similarity measures. We observe that the performance of similarity measures gets affected due to the regular updates of GO. We find that similarity measures are not robust in all conditions, rather they keep their performance quite similar over the evolution of GO in certain conditions. © 2019, Springer Nature Switzerland AG.

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

Impact of the Continuous Evolution of Gene Ontology on Similarity Measures

Demand for transmission using short block length messages has been increased recently in applications including satellite communication, mobile communication, wireless sensor networks, and machine type communications. Unacceptable performance degradation associated with the short block length Turbo codes restricted its use for applications requiring communication with short block length codewords. A novel performance improved reliability-based decoding algorithm for short block length Turbo codes has been formulated and proposed by the authors. The proposed algorithm has a coding gain of 2.45 dB at a BER of 10−3 over AWGN channel with BPSK modulation for a code rate of 1/4. The algorithm has a channel adaptive complexity and has shown nearly 82% reduction in the decoding time complexity for the rate 1/4 Turbo code at 3 dB SNR. As an extension of this earlier work, a detailed performance analysis of the algorithm on different Turbo codes has been carried out. A four-state Turbo encoder has been used to bring out the key aspects of the level based algorithm which is driven by reliability as the key parameter. The formulated algorithm has been applied to different Turbo encoder structures namely 3GPP LTE and CCSDS Turbo codes and a detailed analysis has been carried out in this paper. Simulation results show a significant improvement in the error correction performance of short block length Turbo codes. The algorithm also leads to a marked improvement in time complexity at high SNRs. The algorithm is an attractive solution for applications requiring communication with short block lengths. © 2019, © 2019 IETE.

IETE Journal of Research

Performance Analysis of Reliability-Based Decoding Algorithm for Short Block Length Turbo Codes

Detection of objects in aerial images has gained significant attention in recent years, due to its extensive needs in civilian and military reconnaissance and surveillance applications. With the advent of Unmanned Aerial Vehicles (UAV), the scope of performing such surveillance task has increased. The small size of the objects in aerial images makes it very difficult to detect them. Two-stage Region based Convolutional Neural Network framework for object detection has been proved quite effective. The main problem with these frameworks is the low speed as compared to the one class object detectors due to the computation complexity in generating the region proposals. Region-based methods suffer from poor localization of the objects that leads to a significant number of false positives. This paper aims to provide a solution to the problem faced in real-time vehicle detection in aerial images and videos. The proposed approach used hyper maps generated by skip connected Convolutional network. The hyper feature maps are then passed through region proposal network to generate object like proposals accurately. The issue of detecting objects similar to background is addressed by modifying the loss function of the proposal network. The performance of the proposed network has been evaluated on the publicly available VEDAI dataset. © 2019, Springer Nature Switzerland AG.

Real-Time Vehicle Detection in Aerial Images Using Skip-Connected Convolution Network with Region Proposal Networks

Recognizing text from degraded and low-resolution document images is still an open challenge in the vision community. Existing text recognition systems require a certain resolution and fails if the document is of low-resolution or heavily degraded or noisy. This paper presents an end-to-end trainable deep-learning based framework for joint optimization of document enhancement and recognition. We are using a generative adversarial network (GAN) based framework to perform image denoising followed by deep back projection network (DBPN) for super-resolution and use these super-resolved features to train a bidirectional long short term memory (BLSTM) with Connectionist Temporal Classification (CTC) for recognition of textual sequences. The entire network is end-to-end trainable and we obtain improved results than state-of-the-art for both the image enhancement and document recognition tasks. We demonstrate results on both printed and handwritten degraded document datasets to show the generalization capability of our proposed robust framework. © 2019 IEEE.