27 research outputs found
Recognition with Local Features: the Kernel Recipe
No full textRecent developments in computer vision have shown that local features can provide efficient representations suitable for robust object recognition. Support Vector Machines have been established as powerful learning algorithms with good generalization capabilities. In this paper, we combine these two approaches and propose a general kernel method for recognition with local features. We show that the proposed kernel satisfies the Mercer condition and that this type of kernel is suitable for many standard local feature techniques in computer vision. Large-scale recognition results are presented on three different databases, which demonstrate that SVMs using the proposed kernel perform better than standard Nearest-Neighbor techniques on local features. In addition, experiments on noisy and occluded images show that local feature representations significantly outperform global approaches
Evaluation of Judicial Performance: A Tool for Self-Improvement
No full textThe quality of our judicial system, like other institutions, is a function of the work performed by those who are afforded major roles in the dispensation of justice. Unmistakably. judges, jurors and lawyers assume key roles in this process. Professor Aynes, who is a member of the A.B.A.\u27s Evaluation of Judicial Performance Committee, recognizes that both judges and lawyers, unlike jurors, are professionals expected to bring more to the bench than honesty, good faith and diligence. The author observes that while efforts to improve the daily performance of attorneys have been well under way since the early 1970\u27s, it i now imperative that we address the issue of improving judicial performance. Professor Aynes reviews the historical antecedents of programs evaluating judicial performance, and shows the significant difference between the A.B.A. \u27s proposal and the various surveys and polls that have been previously utilized, while examining the critical issues confronting the profession in designing a responsible system for evaluating judicial performance
Tender Is the Night: Should Your Expert Be?
No full textThis article discusses the practice of tendering an expert for acceptance or certification by the court at trial in the presence of the jury. The article considers Tennessee and Montana state and federal evidence law. The author suggests that Montana courts and lawyers should comply with the A.B.A. Updated Civil Trial Standard 14 and let juries assess the testimony of a Rule 702 witness without a special designation accorded by the judge certifying a witness as an expert in his or her field
Reprogramming astroglia into neurons with hallmarks of fast-spiking parvalbumin-positive interneurons by phospho-site-deficient Ascl1
No full textCellular reprogramming of mammalian glia to an induced neuronal fate holds the potential for restoring diseased brain circuits. While the proneural factor achaete-scute complex-like 1 (Ascl1) is widely used for neuronal reprogramming, in the early postnatal mouse cortex, Ascl1 fails to induce the glia-to-neuron conversion, instead promoting the proliferation of oligodendrocyte progenitor cells (OPC). Since Ascl1 activity is posttranslationally regulated, here, we investigated the consequences of mutating six serine phospho-acceptor sites to alanine (Ascl1SA6) on lineage reprogramming in vivo. Ascl1SA6 exhibited increased neurogenic activity in the glia of the early postnatal mouse cortex, an effect enhanced by coexpression of B cell lymphoma 2 (Bcl2). Genetic fate-mapping revealed that most induced neurons originated from astrocytes, while only a few derived from OPCs. Many Ascl1SA6/Bcl2-induced neurons expressed parvalbumin and were capable of high-frequency action potential firing. Our study demonstrates the authentic conversion of astroglia into neurons featuring subclass hallmarks of cortical interneurons, advancing our scope of engineering neuronal fates in the brain.We are grateful to B. Rico (King’s college london) for support throughout the project. We acknowledge support from the Microscopy core Facility of the institute of Molecular Biology (iMB) Mainz. Funding: this research was funded in part by the Wellcome trust (206410/Z/17/Z). For the purpose of open access, the author has applied a cc BY public copyright license to any author- accepted manuscript version arising from this submission. this study was also supported by funding from the european Research council (eRc) under the european Union’s horizon 2020 Research and innovation Programme (grant agreement no. 101021560, iMAGine), the German Research Foundation (Be 4182/11- 1, project no. 357058359; cRc1080, project no. 221828878), an eRA- net neuron grant (Brain4Sight, 01eW2202), the research initiative of Rheinland- Pfalz at the Johannes Gutenberg University Mainz (ReAlity) to B.B.; by the inneruniversitäre Forschungsförderung Stufe i of the Johannes Gutenberg University Mainz to S.P.; and by core funding to the Francis crick institute from cancer Research United Kingdom, the Medical Research council, and the Wellcome trust (Fc001002). n.M. was supported by a fellowship from the human Frontiers Science Program (hFSP long- term Fellowship, lt000646/2015). F.F.S. was supported by a fellowship from São Paulo Research Foundation (FAPeSP) process no. 2021/13515- 5. c.S. holds the dixon Family chair in Ophthalmology Research. Author contributions: conceptualization: B.B. design of experiments: n.M., S.P., A.B.A., and B.B. Performed and analyzed in vivo experiments: n.M., S.P., A.B.A., c.G., F.F.S., and R.W. Performed and analyzed RnAscope experiments: S.P. and A.B.A. electrophysiology experiments: n.M. Formal analysis: n.M., S.P., and A.B.A. visualization: n.M., S.P., and B.B. Resources: c.S., M.K., and S.G. Supervision: B.B. Writing—original draft: n.M., S.P., A.B.A, and B.B. Writing—review and editing: All authors. Competing interests: Patent application for SB2021- 002 “Method and compositions for neuronal Reprogramming” has been submitted by the Sunnybrook Research institute and the Governing council of the University of toronto [US patent no. 18/568,747, filed 8 december 2023 (pending)]. c.S. is an author of the patent. the other authors declare they have no competing interests. Data and materials availability: the raw data supporting the conclusions of this article will be openly available from the King’s college london research data repository, KORdS, at doi: 10.18742/23911320. Any reagents that this study generated will be shared by the corresponding author upon reasonable request. All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.Peer reviewe
View-based Dynamic Object Recognition Based on Human Perception
No full textPsychophysical studies have shown that humans actively exploit temporal information such as contiguity of images in object recognition.
We have recently developed a recognition system which uses temporal contiguity to learn extensible representations of objects on-line.
The system performs well both on real-world and synthetic data and shows robustness under illumination changes.
In this paper, we present results which compare the proposed representation against simple image-based representations of the same
complexity using Minkowski Minimum Distance classifiers and Support Vector Machine classifiers. Recognition results for all classifiers show large improvements with incorporated
temporal information
The Causes of Regional Sea-level Change: since 1993
No full textAs a result of climate change, sea level is changing all over the world at unprecedented rates. Sea-level change can have significant impacts on coastal communities, infrastructure and global economy, as most of the major cities are located near to or at the coast. Rising sea levels can lead to, for instance, more severe and more frequent flooding, increasing coastal erosion and salt water intrusion. In addition, sea-level change can also influence coastal ecosystems, by altering the habitats of many plant and animals species. Therefore, it is crucial that we understand what is causing sea-level change and at what rate sea levels are changing.Global mean sea level has been rising at a rate of about 3.4 millimetres per year over the last 30 years. Regionally, however, sea level can be changing at a much higher or lower rate. That is because local processes, such as ocean dynamics and gravitational effects associated with continental ice mass changes, cause regional deviations from the global average. But what is causing sea level to change at a specific location? Is sea level changing because the oceans are warming, and thus expanding? Or because the ice from glaciers and ice sheets are melting? The attribution of sea-level change to these and other drivers can be done using a sea-level budget approach. Sea-level budget studies can be used to constrain missing or poorly known contributions and to validate climate models. While the global mean sea-level budget is considered closed within uncertainties, closing the budget on a regional to local scale is still challenging.In this thesis, I focused on the question: Can we close the regional sea-level budget in the satellite altimetry era on a sub-basin scale consistently for the entire world? For this, we need not only high quality observations of sea-level change and each component, but also of the uncertainties within each process. Therefore, in Chapter 2 and 3, I explored the main drivers of regional sea-level change, focusing on the uncertainty characterization of each component. I then looked at which spatial scale is optimal for analysing the regional sea-level budget, and compared the sum of the drivers with the total observed change in these regions in Chapter 4
PAST AND PRESENT LAND TENURE SYSTEMS IN ALBANIA: PATRILINEAL, PATRIARCHAL, FAMILY-CENTERED
No full textThis paper attempts to evaluate whether Albanian rural social structure has changed to the extent that individual rights and protection of those rights have become important policy questions. If the evaluation suggests that rural Albanians retain the set of family-oriented norms and beliefs that are based primarily on patriarchalism and patrilineal inheritance, we must address the following questions: How appropriate is the mixture of western law that emulates individualistic notions of property rights with the customary family-tenure system of rural Albania? What are the likely problems that could emerge during the transition given a potential conflict between family notions of ownership and individual notions of ownership? This paper discusses five broad issues: the contemporary importance of family ownership, the role of the patriarch, the contemporary inheritance procedures, the vulnerability of specific groups of women, and the structure of the Albanian family. Keywords: Land tenure -- Albania Right of property -- Albania Inheritance and succession -- Albania Albania -- Social conditionsLand tenure -- Albania, Right of property -- Albania, Inheritance and succession -- Albania, Albania -- Social conditions, Land Economics/Use,
Income and transfer tax integration: Historic policy links for wealth transfer tax restructuring
No full textThe proposals listed herein are an outgrowth of recent study by this author and Ms. Sharon K. Brougham, M.T., C.P. A., who is a doctoral accounting student at the University of Colorado at Boulder. The scope of this article does not allow for full elaboration so only key highlights of the study are listed. The overall intent is to update prior discussions on estate-income tax unification and to foster further debate as to the efficacy of retaining the present dual-track system of taxation on individuals. It is not, however, intended to be the finite blueprint of tax reform. The full study is scheduled to be published in the Akron Tax Law Journal along with the A.B.A. Report
Improving "bag-of-keypoints" image categorisation: Generative Models and PDF-Kernels
No full textIn this paper we propose two distinct enhancements to the basic
''bag-of-keypoints" image categorisation scheme proposed in [4]. In this
approach images are represented as a variable sized set of local image
features (keypoints). Thus, we require machine learning tools which
can operate on sets of vectors. In [4] this is achieved by representing
the set as a histogram over bins found by k-means. We show how this
approach can be improved and generalised using Gaussian Mixture Models
(GMMs). Alternatively, the set of keypoints can be represented directly
as a probability density function, over which a kernel can be de ned. This
approach is shown to give state of the art categorisation performance
On the performance of coefficient of variation control charts in Phase I
No full textControl charts are mainly carried out in 2 interconnected phases: Phase I (retrospective phase) and Phase II (monitoring phase). Phase I uses a stable historical sample to establish control limits that will be used later in Phase II. The preciseness of the control limits obtained from Phase I can greatly affect the performance of control charts in Phase II. Monitoring the coefficient of variation (CV) is an effective approach when the process mean or standard deviation is not constant. Until now, little work has been dedicated on investigating the performance of CV control charts in Phase I. Viewed under this perspective, this study investigates the performance of CV control charts in Phase I in terms of probability to signal. A real-life example is also provided to illustrate the working of CV charts in Phase I.The author Saddam Akber Abbasi would like to acknowledge the support of Qatar University, Qatar for providing excellent research facilities. The authors Abdaljbbar Dawod and Marwan Al-Momani would like to thank King Fahd University of Petroleum and Minerals, KSA for extraordinary research support.Scopu
