Abstract

Given a monocular color image of a warehouse rack, we aim to predict the bird’s-eye view layout for each shelf in the rack, which we term as ‘multi-layer’ layout prediction. To this end, we present RackLay, a deep neural network for real-time shelf layout estimation from a single image. Unlike previous layout estimation methods which provide a single layout for the dominant ground plane alone, RackLay estimates the top-view and front-view layout for each shelf in the considered rack populated with objects. RackLay’s architecture and its variants are versatile and estimate accurate layouts for diverse scenes characterized by varying number of visible shelves in an image, large range in shelf occupancy factor and varied background clutter. Given the extreme paucity of datasets in this space and the difficulty involved in acquiring real data from warehouses, we additionally release a flexible synthetic dataset generation pipeline WareSynth which allows users to control the generation process and tailor the dataset according to contingent application. The ablations across architectural variants and comparison with strong prior baselines vindicate the efficacy of RackLay as an apt architecture for the novel problem of multi-layered layout estimation. We also show that fusing the top-view and front-view enables 3D reasoning applications such as metric free space estimation for the considered rack.

Abstract

Typical fingerprint recognition systems are comprised of a spoof detection module and a subsequent recognition module, running one after the other. In this paper, we reformulate the workings of a typical fingerprint recognition system. In particular, we posit that both spoof detection and fingerprint recognition are correlated tasks. Therefore, rather than performing the two tasks separately, we propose a joint model for spoof detection and matching to simultaneously perform both tasks without compromising the accuracy of either task. We demonstrate the capability of our joint model to obtain an authentication accuracy (1:1 matching) of TAR = 100% @ FAR = 0.1% on the FVC 2006 DB2A dataset while achieving a spoof detection ACE of 1.44% on the LiveDet 2015 dataset, both maintaining the performance of stand-alone methods. In practice, this reduces the time and memory requirements of the fingerprint recognition system by 50% and 40%, respectively; a significant advantage for recognition systems running on resource-constrained devices and communication channels.

Abstract

There has been increasing interest in building deep hierarchy-aware classifiers that aim to quantify and reduce the severity of mistakes, and not just reduce the number of errors. The idea is to exploit the label hierarchy (e.g., the WordNet ontology) and consider graph distances as a proxy for mistake severity. Surprisingly, on examining mistake-severity distributions of the top-1 prediction, we find that current state-of-the-art hierarchy-aware deep classifiers do not always show practical improvement over the standard cross-entropy baseline in making better mistakes. The reason for the reduction in average mistake-severity can be attributed to the increase in low-severity mistakes, which may also explain the noticeable drop in their accuracy. To this end, we use the classical Conditional Risk Minimization (CRM) framework for hierarchy-aware classification. Given a cost matrix and a reliable estimate of likelihoods (obtained from a trained network), CRM simply amends mistakes at inference time; it needs no extra hyperparameters and requires adding just a few lines of code to the standard cross-entropy baseline. It significantly outperforms the state-of-the-art and consistently obtains large reductions in the average hierarchical distance of top- predictions across datasets, with very little loss in accuracy. CRM, because of its simplicity, can be used with any off-the-shelf trained model that provides reliable likelihood estimates.

Abstract

Information Security in the cyber world is a major cause for concern, with significant increase in the number of attack surfaces. Existing information on vulnerabilities, attacks, controls, and advisories available on the web provides an opportunity to represent knowledge and perform security analytics to mitigate some of the concerns. Representing security knowledge in the form of ontology facilitates anomaly detection, threat intelligence, reasoning and relevance attribution of attacks, and many more. This necessitates dynamic and automated enrichment of information security ontologies. However, existing ontology enrichment algorithms based on natural language processing and ML models have issues with the contextual extraction of concepts in words, phrases and sentences. This motivates the need for sequential Deep Learning architectures that traverse through dependency paths in text and extract embedded vulnerabilities, threats, controls, products and other security related concepts and instances from learned path representations. In the proposed approach, Bidirectional LSTMs trained on a large DBpedia dataset and Wikipedia corpus of 2.8 GB along with Universal Sentence Encoder was deployed to enrich ISO 27001 [8] based information security ontology. The approach yielded a test accuracy of over 80% when tested with knocked out concepts from ontology and web page instances to validate the robustness.

Abstract

Increasing usage of Information Technology (IT) applications in distributed environment is leading to an increase in security exploits. Vulnerabilities related information is also available on open web in an unstructured format that developers may leverage to fix security weaknesses in their IT applications. SQL Injection (SQLi) is one of the topmost vulnerabilities impacting the security of IT applications. We propose an approach to identify information about SQLi in text using text summarization to process any length of text, and a supervised machine learning model to automate the classification of SQLi. To validate the proposed approach, we created a dataset of 100,019 entries that includes 50,010 entries of SQLi from the National Vulnerability Database, 25,010 near negatives related to other cyber security vulnerabilities, and 24,999 data entries that are unrelated to cyber security. The selected Random Forest model was also tested identify SQLi from Web and Twitter text.

Abstract

Earthquakes cause disasters only when the built environment fails to resist the earthquake shaking experienced in the affected area, thereby causing losses of life and property. Ensuring structural safety of the built environment alone will result in earthquake safety of India. India needs suffcient number of competent structural, geotechnical engineers and architects to design and construct its new earthquake resistant buildings and facilities and to retroft its existing ones. This paper urges quantum changes in the educational, social, technical, fnancial, techno-legal, industrial and administrative governance systems of the country, and re-iterates the stand of the nation of zero tolerance to avoidable deaths due to earthquakes.

Abstract

The Koyna-Warna region of Maharashtra, India, is one of the most signifcant worldwide examples for reservoir-induced seismicity. The area is highly vulnerable to earthquakes, and it has experienced over 1 lakh number of shocks since 1963. The largest known earthquake of magnitude 6.5 (Richter scale) occurred on 10th December 1967. Many low and moderate earthquake events have occurred over the past 50 years. A structured survey using rapid visual screening was carried out for existing RC buildings. The seismic risk index depends on three parameters, viz. hazard, exposure, and vulnerability. Many existing RC buildings in the Koyna-Warna region are designed to resist the gravity loads only without any seismic resistant provisions. Hence, there is a need to study the risk index of these RC buildings to assess future serious risks. In this study, the rapid visual survey of 120 existing RC buildings has been done through a modifed EDRI method (Earthquake Disaster Risk Index) to evaluate the seismic risk index of the Koyna-Warna region (Zone-IV as per IS: 1893–2016). The results depict that the risk index of RC buildings in the Koyna-Warna region is in severe damage condition and hence there is a need to take an initiatives for earthquake preparedness plan, with emphasis on retroftting measures, to reduce the loss of human life and damage to physical infrastructure in future seismic events.

Abstract

Search strategies for the third-generation leptoquarks (LQs) are distinct from other LQ searches, especially when they decay to a top quark and a τ lepton. We investigate the cases of all TeV-scale scalar and vector LQs that decay to either a top-tau pair (charge-1/3 and 5/3 LQs) or a top-neutrino pair (charge2/3 LQs). One can then use the boosted top (which can be tagged efficiently using jet-substructure techniques) and high-pT τ leptons to search for these LQs. We consider two search channels with either one or two taus along with at least one hadronically decaying boosted top quark. We estimate the high luminosity LHC (HL-LHC) search prospects of these LQs by considering both symmetric and asymmetric pair and single production processes. Our selection criteria are optimised to retain events from both pair and single production processes. The combined signal has better prospects than the traditional searches. We include new three-body single production processes to enhance the single production contributions to the combined signal. We identify the interference effect that appears in the dominant single production channel of charge-1/3 scalar LQ (S 1/3 ). This interference is constructive if S 1/3 is weak-triplet and destructive, for a singlet one. As a result, their LHC prospects differ appreciably

Abstract

The U1 leptoquark (LQ) is known to be a suitable candidate for explaining the semileptonic B-decay anomalies. We derive precise limits on its parameter space relevant for the anomalies from the current LHC high-pT dilepton data. We consider an exhaustive list of possible B-anomalies-motivated simple scenarios with one or two new couplings that can also be used as templates for obtaining bounds on more complicated scenarios. To obtain precise limits, we systematically consider all possible U1 production processes that can contribute to the dilepton searches including the resonant pair and single productions, nonresonant t-channel U1 exchange, as well as its large interference with the Standard Model background. We demonstrate how the inclusion of generally ignored resonant productions and signal-background interference contributions in the dilepton signal can lead to stringent exclusion limits. We point out new search channels of U1 that can act as unique tests of the flavour-motivated models. The template scenarios can also be used for future U1 searches at the LHC. We compare the LHC limits with other relevant flavor bounds and find that a TeV-scale U1 can accommodate both RD(∗) and RK(∗) anomalies while satisfying all the bounds.

Abstract

The coronavirus disease 2019 (COVID-19), caused by the virus SARS-CoV-2, is an acute respiratory disease that has been classified as a pandemic by the World Health Organization (WHO). The sudden spike in the number of infections and high mortality rates have put immense pressure on the public healthcare systems. Hence, it is crucial to identify the key factors for mortality prediction to optimize patient treatment strategy. Different routine blood test results are widely available compared to other forms of data like X-rays, CT-scans, and ultrasounds for mortality prediction. This study proposes machine learning (ML) methods based on blood tests data to predict COVID-19 mortality risk. A powerful combination of five features: neutrophils, lymphocytes, lactate dehydrogenase (LDH), high-sensitivity C-reactive protein (hs-CRP), and age helps to predict mortality with 96% accuracy. Various ML models (neural networks, logistic regression, XGBoost, random forests, SVM, and decision trees) have been trained and performance compared to determine the model that achieves consistently high accuracy across the days that span the disease. The best performing method using XGBoost feature importance and neural network classification, predicts with an accuracy of 90% as early as 16 days before the outcome. Robust testing with three cases based on days to outcome confirms the strong predictive performance and practicality of the proposed model. A detailed analysis and identification of trends was performed using these key biomarkers to provide useful insights for intuitive application. This study provide solutions that would help accelerate the decision-making process in healthcare systems for focused medical treatments in an accurate, early, and reliable manner