13027 research outputs found
Sort by
A comparison of spherical neural networks for surround-view fisheye image semantic segmentation
No full textThe automotive industry has made significant strides in enhancing road safety and enabling automated driving features through advanced computer vision techniques. This is particularly true for short-range vehicle automation, where non-linear fisheye cameras are commonly used. However, these cameras are challenged by optical distortions, known as fisheye geometric distortions, which lead to object deformation within the image and significant pixel distortion, particularly at the image periphery. Based on the observation that fisheye and spherical images exhibit at least superficially similar geometric characteristics, we investigate the applicability of spherical models—including Spherical Convolutional Neural Networks (CNNs) and Spherical Vision Transformers (ViTs)—to fisheye images, even though fisheye images are not truly spherical. We perform our comparison using fisheye datasets—Woodscape, SynWoodscape, and SynCityscapes in autonomous driving scenarios, with a specific focus on the ability of spherical methods (Spherical CNNs and ViTs) to manage fisheye distortions and compared them against traditional non-spherical methods. Our findings indicate that spherical methods effectively address fisheye distortions without needing extra data augmentations. This results in better mean Intersection over Union (mIoU) scores, pixel accuracy, and better surround-view perception than other modern approaches for fisheye semantic segmentation. However, we also find that spherical methods have a greater tendency to overfit smaller datasets compared with non-spherical models. These advancements highlight how non-linear camera images can take advantage of spherical approximations through spherical models in autonomous driving.</p
Transforming gender equality and inclusion in higher education in Ireland
No full textToday, Ireland is recognised as a leader among European countries in advancing gender equality (GE) in higher education. The remarkable characteristic of this profile, though, is the speed at which the higher education (HE) sector has embedded GE as a distinctive and integral feature of the system. Although there were efforts to advance gender equality in the Irish higher education sector, the importance of the topic did not take hold until 2015-2016. At that point, two events of enduring significance occurred. One was consensus in the higher education institutions (HEIs), supported by the Higher Education Authority (HEA), to adopt the Athena Swan framework for gender equality plans (GEPs). The other was publication by the HEA of an investigative report, with recommendations, by a committee led by former European Commissioner for Research and Innovation, Maire Geoghegan-Quinn.Little is known about advancing change across a higher education system in one country as a national priority. This paper explores the changes within the Irish HE landscape to unearth lessons learned and to address how Ireland as a case study, with government focus and support, has transformed the EDI agenda in the context of GE. The paper is based on, and draws from, a recent publication on this subject to which change leaders have contributed reflections to embed an EDI landscape. It discusses the methods by which collaboration on this book project was developed and supported, and the value of this reflective approach in making sense of the fast-paced change of the last decade. The paper highlights challenges and well as opportunities encountered along the way and briefly reflects on the maturation of the gender equality agenda in the current context</p
Masculine identity in Arab(ic) fiction
No full textThe Arab world is deeply marked by the challenges and controversies surrounding gender and sexuality, in its societies, cultures, politics, and, most notably, literature. This research study endeavours to address perplexing portrayals and dilemmatic representations related to gender complexity and non-normative sexuality of particular male characters in four Arabic novels: Midaq Alley (Zuqaq al-Midaqq); The Stone of Laughter (Hajar al-Dahk); In the Spider’s Room (Fi ghurfat al-ʻankabut); and, exceptionally, Guapa. In dealing with gender and sexuality as main categories of analysis that are socially and culturally constructed over time, according to Judith Butler in Gender Trouble, the tangled concerns of the chosen narratives offer a noteworthy opportunity to critically examine complex yet thought-provoking depictions of these categories at the level of particular characters and their identities. In so doing, the study devotes in-depth attention to the perception of gender through Butler’s prism, along with non-normative sexuality, which are considered controversial topics both in reality and fiction. In respect of this, the thesis features analytical, critical, and close readings of these novels that are meant to be read, specifically in relation to certain male characters as mentioned above, from Butler’s gender performativity in their venture to challenge essentialist perceptions not only of gender but also of sex and sexuality. This directs the researcher to examine how the Arabic novel broaches and exposes these problematics, shedding light on the (un)doing of masculine gender identity, along with masculinity, as well as the subversion of sexuality. Regarding this, even though addressing homosexuality in Arabic literature has been a notable concern in previous research, dealing with this corner as a conspicuous dilemma in conjunction with gender still represents a critical gap in the Arabic novel. As such, researchers in the field of Arabic literature in relation to gender and sexuality studies have not addressed subversive portrayals of gender and sexuality in a holistic literary and theoretical framework. In addition, Butler’s claims on the (de)construction of gender and sexuality have not yet been implemented and illustrated through an Arabic literary lens.Above that, dealing with gender as a doing, as interpreted by Butler themselves, has not yet been studied in an Arabic novel hand in hand with homosexuality. Thus, the objective is to fill these voids, highlighting the sensitivity of such taboo matters in the Arabic novel. Ultimately, this research will contribute to a comprehensive examination of these representations. It will unveil the role played by the Arabic novel and its authors in broaching these corners, despite their controversy in the Arab world and its socio-political milieus, as portrayed throughout the chosen narratives.</p
DriVQA: A gaze-based dataset for visual question answering in driving scenarios
No full textThis paper presents DriVQA, a novel dataset that combines gaze plots and heatmaps with visual question-answering (VQA) data from participants who were presented with driving scenarios. Visual Questioning Answering (VQA) is proposed as a part of the vehicle autonomy trustworthiness and interpretability solution in decision-making by autonomous vehicles. Collected using the Tobii Pro X3-120 eye-tracking device, the DriVQA dataset provides a comprehensive mapping of where participants direct their gaze when presented with images of driving scenes, followed by related questions and answers from every participant. For each scenario, the dataset contains: images of driving situations, associated questions, participant answers, gaze plots, and heatmaps. It is being used to study the subjectivity inherent in VQA. Its detailed gaze-tracking data offers a unique perspective on how individuals perceive and interpret visual scenes, making it an essential resource for training VQA models that rely on human-like attention. The dataset is a valuable tool for investigating human cognition and behaviour in dynamic, real-world scenarios. DriVQA is highly relevant for VQA models, as it allows the systems to learn from human-like attention behaviour when making decisions based on visual input when trained. The dataset has the potential to drive advancements in VQA research and development by improving the safety and intelligence of driving systems through enhanced visual understanding and interaction. DriVQA has significant potential for reuse in various research areas, including the development of advanced VQA models, attention analysis, and human-computer interaction studies. Its comprehensive gaze plots and heatmaps can also be leveraged to improve applications in autonomous driving, driver assistance systems, and cognitive science research, making it a versatile resource for both academic and industrial purposes.</p
Functionalized organosolv lignin grafted with 3-aminopropyltrie-thoxysilane: A bio-based adsorbent for phosphate recovery from dairy wastewater
No full textWastewater rich in phosphates and nitrates causes eutrophication and leads to the impairment of freshwater resources. Out of various methods used, adsorption is the immaculate and economical for removing and recovering phosphates and nitrates from wastewater streams in a single-step process. This study explores the potential of bio-based adsorbent, functionalized organosolv lignin [OL], chemically modified by grafting with 3-aminopropyltriethoxysilane [APTES], as an effective bio-based adsorbent [OL-APTES-H+] for phosphate recovery from aqueous solutions and industrial wastewater. The characterization of OL-APTES-H+ was performed using multiple analytical techniques, providing comprehensive information on the material morphology, elemental composition, functional groups, thermal stability, surface charge, and electrokinetic behavior. The adsorption efficiency of OL-APTES-H+ was assessed under varying experimental conditions, including pH, contact time, and initial phosphate concentration. The adsorption capacity of OL-APTES-H+ depended on pH, with different forms of phosphate species being preferentially adsorbed at different pH values. A maximum adsorption capacity of 21.12 mg/g was achieved at pH 5. Kinetic studies indicated that the adsorption process followed a combination of electrostatic interactions, chemisorption and surface interaction, as evidenced by SEM and EDS analyses. XPS results confirm phosphorus incorporation on the adsorbent surface, reinforcing chemisorption. Adsorption isotherm analysis revealed that the data fitted well to the Langmuir isotherm model, suggesting a monolayer adsorption mechanism. The adsorption performance of OL-APTES-H+ was enhanced in the presence of monovalent ions, while a slight reduction in efficiency was observed in the presence of divalent anions. When applied to industrial dairy wastewater, OL-APTES-H+ exhibited phosphate removal efficiencies ranging from 30 % to 58 %. Overall, OL-APTES-H+ demonstrates considerable potential as a bio-based adsorbent for phosphate recovery, effectively mitigating environmental pollution in wastewater bodies and providing an eco-friendly source of phosphates for sustainable agricultural practices.</p
Intangible income shifting: the impact of digitalisation on falling corporate tax rates
No full textConcerns over the amount of corporation tax paid by firms in digitally intensive industries have led to efforts to reform corporate tax codes at national and international levels. Despite this, there is little evidence to suggest firms in digitally intensive industries pay less corporate tax. The paper examines the interplay of digitalisation, firm size and MNE status to investigate if large, digitally intensive and multinational firms do indeed pay less tax, as measured by the cash effective tax rate (Cash ETR). It finds no significant relationship between industry digitalisation and Cash ETRs using a primary measure of industry digitalisation. However, the inter-actions of digitalisation, firm size and MNE status exhibit statistically significant impacts on Cash ETRs, though the magnitudes are modest. Larger firms in digital intensive industries tend to have Cash ETRs that are 1% lower than their less digital intensive equivalents. MNEs in more digital intensive industries tend to have Cash ETRs that are approximately 0.5% lower than firms in less digital intensive industries. Notably, the very largest tech firms display considerably lower Cash ETRs, thus, it would appear they provide much of the impetus for corporate tax reform.</p
Illuminating the dark kinome: utilizing multiplex peptide activity arrays to functionally annotate understudied kinases
No full textProtein kinases are critical components of a myriad biological processes and strongly associated with various dis?eases. While kinase research has been a point of focus in biomedical research for several decades, a large portion of the kinome is still considered understudied or“dark,” because prior research is targeted towards a subset of kinases with well-established roles in cellular processes. We present an empirical and in-silico hybrid workfow to extend the functional knowledge of understudied kinases. Utilizing multiplex peptide activity arrays and robust in-silico analyses, we extended the functional knowledge of fve dark tyrosine kinases (AATK, EPHA6, INSRR, LTK, TNK1) and explored their roles in schizophrenia, Alzheimer’s dementia (AD), and major depressive disorder (MDD). Using this hybrid approach, we identifed 195 novel kinase-substrate interactions with variable degrees of afnity and linked extended functional networks for these kinases to biological processes that are impaired in psychiatric and neuro?logical disorders. Biochemical assays and mass spectrometry were used to confrm a putative substrate of EPHA6, an understudied dark tyrosine kinase. We examined the EPHA6 network and knowledgebase in schizophrenia using reporter peptides identifed and validated from the multi-plex array with high afnity for phosphorylation by EPHA6. Identifcation and confrmation of putative substrates for understudied kinases provides a wealth of actionable infor?mation for the development of new drug treatments as well as exploration of the pathophysiology of disease states using signaling network approaches.</p
Recent advances in the postbuckling analysis of composite laminated structures
No full textThe current demand for lightweight structures in a wide range of engineering applications leads to using thin-walled composite laminated structures whose behaviour is governed by buckling and postbuckling phenomena. Such a demand is pushing the borders of computational mechanics to enhance methods and algorithms for studying those structures' geometrically nonlinear responses. This work presents some of the authors' developments in analysing lightweight composite laminated structures. The literature survey introduces a family of finite elements known as MISS elements, where MISS stands for mixed isostatic self-equilibrated stresses. The description of those elements, which are derived from the Hellinger-Reissner functional, is followed by a discussion on their advantages concerning displacement-based elements when studying composite laminated thin-walled structures. Subsequently, a framework for the postbuckling analysis of composite structures with the MISS-4C element that uses the Koiter multi-modal approach is presented.</p
Dynamics of liquid infiltration into an espresso bed using time-resolved micro-computed tomography: Insights from experiment and modeling
No full textIn espresso brewing, soluble content is extracted from a bed of ground coffee beans by forcing hot water through the bed at high pressure. An important part of this process is the infiltration stage in which water permeates the initially dry bed. This process is investigated by a combination of x-ray tomography and fluid mechanical modeling. Tomography is used to track the infiltration front of the water via the contrast in density. The experimental data are compared with a one-dimensional unsaturated porous medium flow model, which divides the bed into wet and dry regions and incorporates the espresso pump dynamics. Good agreement is seen between the experimental data and the model predictions</p