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

World University root node

Hybrid Sample Synthesis-based Debiasing of Classifier in Limited Data Setting

Single-cell RNA-sequencing (scRNA-seq) offers unparalleled insight into the transcriptional programs of different cellular states by measuring the transcriptome of thousands of individual cells. An emerging problem in the analysis of scRNA-seq is the inference of transcriptional gene regulatory networks and a number of methods with different learning frameworks have been developed to address this problem. Here, we present an expanded benchmarking study of eleven recent network inference methods on seven published scRNA-seq datasets in human, mouse, and yeast considering different types of gold standard networks and evaluation metrics. We evaluate methods based on their computing requirements as well as on their ability to recover the network structure. We find that, while most methods have a modest recovery of experimentally derived interactions based on global metrics such as Area Under the Precision Recall curve, methods are able to capture targets of regulators that are relevant to the system under study. Among the top performing methods that use only expression were SCENIC, PIDC, MERLIN or Correlation. Addition of prior biological knowledge and the estimation of transcription factor activities resulted in the best overall performance with the Inferelator and MERLIN methods that use prior knowledge outperforming methods that use expression alone. We found that imputation for network inference did not improve network inference accuracy and could be detrimental. Comparisons of inferred networks for comparable bulk conditions showed that the networks inferred from scRNA-seq datasets are often better or at par with the networks inferred from bulk datasets. Our analysis should be beneficial in selecting methods for network inference. At the same time, this highlights the need for improved methods and better gold standards for regulatory network inference from scRNAseq datasets. © The Author(s) 2023. Published by Oxford University Press on behalf of the Genetics Society of America.

G3: Genes, Genomes, Genetics

Identifying strengths and weaknesses of methods for computational network inference from single-cell RNA-seq data

Cell type-specific gene expression patterns are outputs of transcriptional gene regulatory networks (GRNs) that connect transcription factors and signaling proteins to target genes. Single-cell technologies such as single cell RNA-sequencing (scRNA-seq) and single cell Assay for Transposase-Accessible Chromatin using sequencing (scATAC-seq), can examine cell-type specific gene regulation at unprecedented detail. However, current approaches to infer cell type-specific GRNs are limited in their ability to integrate scRNA-seq and scATAC-seq measurements and to model network dynamics on a cell lineage. To address this challenge, we have developed single-cell Multi-Task Network Inference (scMTNI), a multi-task learning framework to infer the GRN for each cell type on a lineage from scRNA-seq and scATAC-seq data. Using simulated and real datasets, we show that scMTNI is a broadly applicable framework for linear and branching lineages that accurately infers GRN dynamics and identifies key regulators of fate transitions for diverse processes such as cellular reprogramming and differentiation. © 2023, The Author(s).

Fulltext

Nature Communications

Inference of cell type-specific gene regulatory networks on cell lineages from single cell omic datasets

Neural Networks

Mitigate forgetting in few-shot class-incremental learning using different image views

Recent advancements in deep learning have enabled 3D human body reconstruction from a monocular image, which has broad applications in multiple domains. In this paper, we propose SHARP (SHape Aware Reconstruction of People in loose clothing), a novel end-to-end trainable network that accurately recovers the 3D geometry and appearance of humans in loose clothing from a monocular image. SHARP uses a sparse and efficient fusion strategy to combine parametric body prior with a non-parametric 2D representation of clothed humans. The parametric body prior enforces geometrical consistency on the body shape and pose, while the non-parametric representation models loose clothing and handles self-occlusions as well. We also leverage the sparseness of the non-parametric representation for faster training of our network while using losses on 2D maps. Another key contribution is 3DHumans, our new life-like dataset of 3D human body scans with rich geometrical and textural details. We evaluate SHARP on 3DHumans and other publicly available datasets, and show superior qualitative and quantitative performance than existing state-of-the-art methods. © 2022, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.

International Journal of Computer Vision

SHARP: Shape-Aware Reconstruction of People in Loose Clothing

Face possesses a rich spatial structure that can provide valuable cues to guide various face-related tasks. The eyes are considered an important socio-visual cue for effective communication. They are an integral feature of facial expressions as they are an important aspect of interpersonal communication. However, virtual reality headsets occlude a significant portion of the face and restrict the visibility of certain facial features, particularly the eye region. Reproducing this region with realistic content and handling complex eye movements such as blinks is challenging. Previous facial inpainting methods are not capable enough to capture subtle eye movements. In view of this, we propose a working solution to refine the reconstructions, particularly around the eye region, by leveraging inherent eye structure. We introduce spatial supervision and a novel landmark predictor module to regularize per-frame reconstructions obtained from an existing image-based facial de-occlusion network. Experiments verify the usefulness of our approach in enhancing the quality of reconstructions to capture subtle eye movements. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

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

Supervision by Landmarks: An Enhanced Facial De-occlusion Network for VR-Based Applications

Leveraging joint incremental learning objective with data ensemble for class incremental learning

With increasing number of COVID-19 cases globally, all the countries are ramping up the testing numbers. While the RT-PCR kits are available in sufficient quantity in several countries, others are facing challenges with limited availability of testing kits and processing centers in remote areas. This has motivated researchers to find alternate methods of testing which are reliable, easily accessible and faster. Chest X-Ray is one of the modalities that is gaining acceptance as a screening modality. Towards this direction, the paper has two primary contributions. Firstly, we present the COVID-19 Multi-Task Network (COMiT-Net) which is an automated end-to-end network for COVID-19 screening. The proposed network not only predicts whether the CXR has COVID-19 features present or not, it also performs semantic segmentation of the regions of interest to make the model explainable. Secondly, with the help of medical professionals, we manually annotate the lung regions and semantic segmentation of COVID19 symptoms in CXRs taken from the ChestXray-14, CheXpert and a consolidated COVID-19 dataset. These annotations will be released to the research community. Experiments performed with more than 2500 frontal CXR images show that at 90% specificity, the proposed COMiT-Net yields 96.80% sensitivity.

Pattern Recognition

Multi-task driven explainable diagnosis of COVID-19 using chest X-ray images

For a positive integer c, a graph G is said to be c-closed if every pair of non-adjacent vertices in G have at most c - 1 neighbours in common. The closure of a graph G, denoted by cl(G), is the least positive integer c for which G is c-closed. The class of c-closed graphs was introduced by Fox et al. [ICALP ‘18 and SICOMP ‘20]. Koana et al. [ESA ‘20] started the study of using cl(G) as an additional structural parameter to design kernels for problems that are W-hard under standard parameterizations. In particular, they studied problems such as Independent Set, Induced Matching, Irredundant Set and (Threshold) Dominating Set, and showed that each of these problems admits a polynomial kernel, either w.r.t. the parameter k + c or w.r.t. the parameter k for each fixed value of c. Here, k is the solution size and c = cl(G). The work of Koana et al. left several questions open, one of which was whether the Perfect Code problem admits a fixed-parameter tractable (FPT) algorithm and a polynomial kernel on c-closed graphs. In this paper, among other results, we answer this question in the affirmative. Inspired by the FPT algorithm for Perfect Code, we further explore two more domination problems on the graphs of bounded closure. The other problems that we study are Connected Dominating Set and Partial Dominating Set. We show that Perfect Code and Connected Dominating Set are fixed-parameter tractable w.r.t. the parameter k + cl(G), whereas Partial Dominating Set, parameterized by k is W[1]-hard even when cl(G) = 2. We also show that for each fixed c, Perfect Code admits a polynomial kernel on the class of c-closed graphs. And we observe that Connected Dominating Set has no polynomial kernel even on 2-closed graphs, unless NP ⊆ co-NP/poly. © Lawqueen Kanesh, Jayakrishnan Madathil, Sanjukta Roy, Abhishek Sahu, and Saket Saurabh.

Leibniz International Proceedings in Informatics, LIPIcs

Further Exploiting c-Closure for FPT Algorithms and Kernels for Domination Problems

In this paper, we address the problem of monocular 3D human reconstruction with an acute focus on the challenge of recovering person-specific facial geometry as well as suppressing surface noise, specifically addressing the issue of false geometrical variations caused by textural edges. Most of the existing state-of-the-art methods in this domain fail to address these challenges. More specifically, we propose to integrate facial and wrinkle map priors in a learning-based framework to improve the quality of full-body 3D reconstruction from monocular images. By incorporating facial prior, we recover person-specific identity unlike many of the existing methods which rely on parametric shape models. Similarly, the wrinkle map prior enables our network to alleviate the challenge of false geometrical variations caused by high-frequency textural details present in the input image. We evaluate our method on publicly available datasets & in-the-wild internet images with loose clothing and report superior performance both qualitatively and quantitatively when compared with SOTA methods. © 2022 ACM.