搜索结果：proposal

Part I. Writing the Proposal Chapter 1. Function of the Proposal Chapter 2. Doing the Right Thing: The Habit of Truth Chapter 3. Developing the Thesis or Dissertation Proposal: Some Common Problems Chapter 4. Content of the Proposal: Important Considerations Chapter 5. Preparation of Proposals for Qualitative Research: Different Assumptions Chapter 6. Proposals for Mixed Methods Research Chapter 7. Style and Form in Writing the Proposal Chapter 8. Oral Presentation Part II. Money for Research Chapter 9. Money for Research: How to Ask for Help Chapter 10. Preparation of the Grant Proposal Part III. Specimen Proposals Proposal 1. Experimental Study Proposal 2. Qualitative Study Proposal 3. Online Electronic Survey Study Proposal 4. Funded Grant

State-of-the-art object detection networks depend on region proposal algorithms to hypothesize object locations. Advances like SPPnet [1] and Fast R-CNN [2] have reduced the running time of these detection networks, exposing region proposal computation as a bottleneck. In this work, we introduce a Region Proposal Network (RPN) that shares full-image convolutional features with the detection network, thus enabling nearly cost-free region proposals. An RPN is a fully convolutional network that simultaneously predicts object bounds and objectness scores at each position. The RPN is trained end-to-end to generate high-quality region proposals, which are used by Fast R-CNN for detection. We further merge RPN and Fast R-CNN into a single network by sharing their convolutional features-using the recently popular terminology of neural networks with 'attention' mechanisms, the RPN component tells the unified network where to look. For the very deep VGG-16 model [3] , our detection system has a frame rate of 5 fps (including all steps) on a GPU, while achieving state-of-the-art object detection accuracy on PASCAL VOC 2007, 2012, and MS COCO datasets with only 300 proposals per image. In ILSVRC and COCO 2015 competitions, Faster R-CNN and RPN are the foundations of the 1st-place winning entries in several tracks. Code has been made publicly available.

Current top performing object detectors employ detection proposals to guide the search for objects, thereby avoiding exhaustive sliding window search across images. Despite the popularity and widespread use of detection proposals, it is unclear which trade-offs are made when using them during object detection. We provide an in-depth analysis of twelve proposal methods along with four baselines regarding proposal repeatability, ground truth annotation recall on PASCAL, ImageNet, and MS COCO, and their impact on DPM, R-CNN, and Fast R-CNN detection performance. Our analysis shows that for object detection improving proposal localisation accuracy is as important as improving recall. We introduce a novel metric, the average recall (AR), which rewards both high recall and good localisation and correlates surprisingly well with detection performance. Our findings show common strengths and weaknesses of existing methods, and provide insights and metrics for selecting and tuning proposal methods.

In this paper, we propose PointRCNN for 3D object detection from raw point cloud. The whole framework is composed of two stages: stage-1 for the bottom-up 3D proposal generation and stage-2 for refining proposals in the canonical coordinates to obtain the final detection results. Instead of generating proposals from RGB image or projecting point cloud to bird's view or voxels as previous methods do, our stage-1 sub-network directly generates a small number of high-quality 3D proposals from point cloud in a bottom-up manner via segmenting the point cloud of the whole scene into foreground points and background. The stage-2 sub-network transforms the pooled points of each proposal to canonical coordinates to learn better local spatial features, which is combined with global semantic features of each point learned in stage-1 for accurate box refinement and confidence prediction. Extensive experiments on the 3D detection benchmark of KITTI dataset show that our proposed architecture outperforms state-of-the-art methods with remarkable margins by using only point cloud as input. The code is available at https://github.com/sshaoshuai/PointRCNN.

This paper introduces a novel rotation-based framework for arbitrary-oriented text detection in natural scene images. We present the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Rotation Region Proposal Networks</i> , which are designed to generate inclined proposals with text orientation angle information. The angle information is then adapted for bounding box regression to make the proposals more accurately fit into the text region in terms of the orientation. The <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Rotation Region-of-Interest</i> pooling layer is proposed to project arbitrary-oriented proposals to a feature map for a text region classifier. The whole framework is built upon a region-proposal-based architecture, which ensures the computational efficiency of the arbitrary-oriented text detection compared with previous text detection systems. We conduct experiments using the rotation-based framework on three real-world scene text detection datasets and demonstrate its superiority in terms of effectiveness and efficiency over previous approaches.

Visual object tracking has been a fundamental topic in recent years and many deep learning based trackers have achieved state-of-the-art performance on multiple benchmarks. However, most of these trackers can hardly get top performance with real-time speed. In this paper, we propose the Siamese region proposal network (Siamese-RPN) which is end-to-end trained off-line with large-scale image pairs. Specifically, it consists of Siamese subnetwork for feature extraction and region proposal subnetwork including the classification branch and regression branch. In the inference phase, the proposed framework is formulated as a local one-shot detection task. We can pre-compute the template branch of the Siamese subnetwork and formulate the correlation layers as trivial convolution layers to perform online tracking. Benefit from the proposal refinement, traditional multi-scale test and online fine-tuning can be discarded. The Siamese-RPN runs at 160 FPS while achieving leading performance in VOT2015, VOT2016 and VOT2017 real-time challenges.

We present AVOD, an Aggregate View Object Detection network for autonomous driving scenarios. The proposed neural network architecture uses LIDAR point clouds and RGB images to generate features that are shared by two subnetworks: a region proposal network (RPN) and a second stage detector network. The proposed RPN uses a novel architecture capable of performing multimodal feature fusion on high resolution feature maps to generate reliable 3D object proposals for multiple object classes in road scenes. Using these proposals, the second stage detection network performs accurate oriented 3D bounding box regression and category classification to predict the extents, orientation, and classification of objects in 3D space. Our proposed architecture is shown to produce state of the art results on the KITTI 3D object detection benchmark [1] while running in real time with a low memory footprint, making it a suitable candidate for deployment on autonomous vehicles. Code is available at: https://github.com/kujason/avod.

The need for standards in the management of patients with endocrine tumors of the digestive system prompted the European Neuroendocrine Tumor Society (ENETS) to organize a first Consensus Conference, which was held in Frascati (Rome) and was based on the recently published ENETS guidelines on the diagnosis and treatment of digestive neuroendocrine tumors (NET). Here, we report the tumor-node-metastasis proposal for foregut NETs of the stomach, duodenum, and pancreas that was designed, discussed, and consensually approved at this conference. In addition, we report the proposal for a working formulation for the grading of digestive NETs based on mitotic count and Ki-67 index. This proposal, which needs to be validated, is meant to help clinicians in the stratification, treatment, and follow-up of patients.

International standardization and coordination of the nomenclature of variants of hepatitis C virus (HCV) is increasingly needed as more is discovered about the scale of HCV-related liver disease and important biological and antigenic differences that exist between variants. A group of scientists expert in the field of HCV genetic variability, and those involved in development of HCV sequence databases, the Hepatitis Virus Database (Japan), euHCVdb (France), and Los Alamos (United States), met to re-examine the status of HCV genotype nomenclature, resolve conflicting genotype or subtype names among described variants of HCV, and draw up revised criteria for the assignment of new genotypes as they are discovered in the future. A comprehensive listing of all currently classified variants of HCV incorporates a number of agreed genotype and subtype name re-assignments to create consistency in nomenclature. The paper also contains consensus proposals for the classification of new variants into genotypes and subtypes, which recognizes and incorporates new knowledge of HCV genetic diversity and epidemiology. A proposal was made that HCV variants be classified into 6 genotypes (representing the 6 genetic groups defined by phylogenetic analysis). Subtype name assignment will be either confirmed or provisional, depending on the availability of complete or partial nucleotide sequence data, or remain unassigned where fewer than 3 examples of a new subtype have been described. In conclusion, these proposals provide the framework by which the HCV databases store and provide access to data on HCV, which will internationally coordinate the assignment of new genotypes and subtypes in the future.

We present Sparse R-CNN, a purely sparse method for object detection in images. Existing works on object detection heavily rely on dense object candidates, such as k anchor boxes pre-defined on all grids of image feature map of size H × W. In our method, however, a fixed sparse set of learned object proposals, total length of N, are provided to object recognition head to perform classification and location. By eliminating HWk (up to hundreds of thousands) hand-designed object candidates to N (e.g. 100) learnable proposals, Sparse R-CNN completely avoids all efforts related to object candidates design and many-to-one label assignment. More importantly, final predictions are directly output without non-maximum suppression post-procedure. Sparse R-CNN demonstrates accuracy, run-time and training convergence performance on par with the well-established detector baselines on the challenging COCO dataset, e.g., achieving 45.0 AP in standard 3× training schedule and running at 22 fps using ResNet-50 FPN model. We hope our work could inspire re-thinking the convention of dense prior in object detectors. The code is available at: https://github.com/PeizeSun/SparseR-CNN.

A uniform system of classification and nomenclature of the acute leukaemias, at present lacking, should permit more accurate recording of the distribution of cases entered into clinical trials, and could provide a reference standard when newly developed cell-surface markers believed to characterize specific cell types are applied to cases of acute leukaemia. Proposals based on conventional morphological and cytochemical methods are offered following the study of peripheral blood and bone-marrow films from some 200 cases of acute leukaemia by a group of seven French, American and British haematologists. The slides were examined first independently, and then by the group working together. Two groups of acute leukaemia, 'lymphoblastic' and myeloid are further subdivided into three and six groups. Dysmyelopoietic syndromes that may be confused with acute myeloid leukaemia are also considered. Photomicrographs of each of the named conditions are presented.

OBJECTIVE: Because of the pressure for timely, informed decisions in public health and clinical practice and the explosion of information in the scientific literature, research results must be synthesized. Meta-analyses are increasingly used to address this problem, and they often evaluate observational studies. A workshop was held in Atlanta, Ga, in April 1997, to examine the reporting of meta-analyses of observational studies and to make recommendations to aid authors, reviewers, editors, and readers. PARTICIPANTS: Twenty-seven participants were selected by a steering committee, based on expertise in clinical practice, trials, statistics, epidemiology, social sciences, and biomedical editing. Deliberations of the workshop were open to other interested scientists. Funding for this activity was provided by the Centers for Disease Control and Prevention. EVIDENCE: We conducted a systematic review of the published literature on the conduct and reporting of meta-analyses in observational studies using MEDLINE, Educational Research Information Center (ERIC), PsycLIT, and the Current Index to Statistics. We also examined reference lists of the 32 studies retrieved and contacted experts in the field. Participants were assigned to small-group discussions on the subjects of bias, searching and abstracting, heterogeneity, study categorization, and statistical methods. CONSENSUS PROCESS: From the material presented at the workshop, the authors developed a checklist summarizing recommendations for reporting meta-analyses of observational studies. The checklist and supporting evidence were circulated to all conference attendees and additional experts. All suggestions for revisions were addressed. CONCLUSIONS: The proposed checklist contains specifications for reporting of meta-analyses of observational studies in epidemiology, including background, search strategy, methods, results, discussion, and conclusion. Use of the checklist should improve the usefulness of meta-analyses for authors, reviewers, editors, readers, and decision makers. An evaluation plan is suggested and research areas are explored.

暂无摘要（点击查看原文获取完整内容）

This paper introduces a proposal to extend the heuristic called "particle swarm optimization" (PSO) to deal with multiobjective optimization problems. Our approach uses the concept of Pareto dominance to determine the flight direction of a particle and it maintains previously found nondominated vectors in a global repository that is later used by other particles to guide their own flight. The approach is validated using several standard test functions from the specialized literature. Our results indicate that our approach is highly competitive with current evolutionary multiobjective optimization techniques.

The observed fit of bone mass to a healthy animal's typical mechanical usage indicates some mechanism or mechanisms monitor that usage and control the three longitudinal growth, bone modeling, and BMU-based remodeling activities that directly determine bone mass. That mechanism could be named a mechanostat. Accumulated evidence suggests it includes the bone itself, plus mechanisms that transform its mechanical usage into appropriate signals, plus other mechanisms that detect those signals and then direct the above three biologic activities. In vivo studies have shown that bone strains in or above the 1500-3000 microstrain range cause bone modelling to increase cortical bone mass, while strains below the 100-300 microstrain range release BMU-based remodeling which then removes existing cortical-endosteal and trabecular bone. That arrangement provides a dual system in which bone modeling would adapt bone mass to gross overloading, while BMU-based remodeling would adapt bone mass to gross underloading, and the above strain ranges would be the approximate "setpoints" of those responses. The anatomical distribution of those mechanical usage effects are well known. If circulating agents or disease changed the effective setpoints of those responses their bone mass effects should copy the anatomical distribution of the mechanical usage effects. That seems to be the case for many agents and diseases, and several examples are discussed, including postmenopausal osteoporosis, fluoride effects, bone loss in orbit, and osteogenesis imperfecta. The mechanostat proposal is a seminal idea which fits diverse evidence but it requires critique and experimental study.

9720384). We greatly appreciate their support but emphasize that the views expressed herein are those of the authors and not of the granting agency. We are very grateful to Carl Bereiter, Asghar (Ali) Iran-Nejad, and David Pearson for excellent comments and suggestions. In addition, we thank our colleagues Kay Burgess, Xiadong Lin, and Sean Brophy for graciously allowing us to discuss some of their yet-to-be-published data. We also thank the members of the Cognition and Technology Group at Vanderbilt who provided invaluable feedback on our work. 2Rethinking Transfer: A Simple Proposal With Multiple Implications A belief in transfer lies at the heart of our educational system. Most educators want learning activities to have positive effects that extend beyond the exact conditions of initial learning. They are hopeful that students will show evidence of transfer in a variety of situations; for example, from one problem to another within a course; from one course to another; from one school year to the next; and from their years in school to their years in the workplace. Beliefs about transfer often accompany the claim that it is better to “educate ” people broadly than simply “train ” them to perform particular tasks (e.g., Broudy, 1977).

T HE HISTOLOGIC categorization of lymphoma has been a source of frustration for many years for both clinicians and pathologists.In the last 10 years, much new information has become available about the lymphomas, resulting in recognition of new entities and refinement of previously recognized disease categories, raising the question of whether it is time for a new lymphoma classification.In this paper we report the result of an international review of lymphomas, which we hope may clarify some of the confusion surrounding this topic.This review was conducted at a meeting of 19 hematopathologists with particular interest and experience in lymphomas (the International Lymphoma Study Group) in Berlin, Germany, in April 1993.At previous meetings in Europe and the United States, we had come to believe that, despite the variety of classification schemes used, many hematopathologists appeared to agree on a rather large number of distinct lymphoma entities that they recognize and diagnose in daily practice.We believed that we could provide a useful service to both pathologists and clinicians struggling with the classification of lymphomas by attempting to arrive at a consensus regarding the categories of lymphoid neoplasia that can be reliably recognized at present.What emerged from this meeting was, first, that each of us had independently evolved ways of viewing these diseases that were essentially identical.Surprisingly, there was little divergence between European and US participants.Second, it was evident that, while many of these lymphoma entities are recognized in the Kiel Classification,"6 the Lukes-Collins Classification,' and the Working Formulation,* they often go by different names in different publications and may have variable criteria for diagno~is.~Furthermore, we found that many of us had doubts about both the practical feasibility and the scientific validity of distinguishing certain subtypes in these systems.We also found that while some lymphoma categories are easy to recognize, others are disturbingly prone to subjective variability.This feature of lymphoma diagnosis has not been emphasized in previous schemes for classification, which imply that all categories are equally easy for the pathologist to recognize.Ideally lymphomas, like most other tumors, should be classified according to their presumed normal counterpart, to the extent possible.This should provide the best information about disease biology, natural history, and response to treatment.However, despite extensive study, the definition of lymphoid compartments in humans and movement of cells between these compartments still contains many uncertainties.Furthermore, there are difficulties in defining the full extent of the neoplastic clone in individual cases of lymphoma, and some well-defined lymphoma types lack obvious normal counterparts.Consequently, although differentiation schemes provide useful conceptual frameworks for

Principles and terminology for classification of the Quaternary are discussed, including lithostratigraphy, biostratigraphy. morphostratigraphy, climatostratigraphy and chronostratigraphy. The main conclusion is a proposal for a common chronostratigraphical classification of the Quaternary in Norden (and partly continental NW Europe). The Quaternary is subdivided into the Pleistocene and the Holocene Series. The Pleistocene is further subdivided into several provisional stages (Weichselian, Eemian, etc.), based on the sequence of glacials/interglacials. but with the boundaries preferably defined by stratotypes. The Late Weichselian and the Flandrian (Holocene) are subdivided into chronozoncs (Bolling, Older Dryas, Allerød, Younger Dryas, Preboreal, Boreal, Atlantic, Subboreal, Subatlantic) with the boundaries dcfined in conventional radiocarbon years.

Molecular structures and sequences are generally more revealing of evolutionary relationships than are classical phenotypes (particularly so among microorganisms). Consequently, the basis for the definition of taxa has progressively shifted from the organismal to the cellular to the molecular level. Molecular comparisons show that life on this planet divides into three primary groupings, commonly known as the eubacteria, the archaebacteria, and the eukaryotes. The three are very dissimilar, the differences that separate them being of a more profound nature than the differences that separate typical kingdoms, such as animals and plants. Unfortunately, neither of the conventionally accepted views of the natural relationships among living systems--i.e., the five-kingdom taxonomy or the eukaryote-prokaryote dichotomy--reflects this primary tripartite division of the living world. To remedy this situation we propose that a formal system of organisms be established in which above the level of kingdom there exists a new taxon called a "domain." Life on this planet would then be seen as comprising three domains, the Bacteria, the Archaea, and the Eucarya, each containing two or more kingdoms. (The Eucarya, for example, contain Animalia, Plantae, Fungi, and a number of others yet to be defined). Although taxonomic structure within the Bacteria and Eucarya is not treated herein, Archaea is formally subdivided into the two kingdoms Euryarchaeota (encompassing the methanogens and their phenotypically diverse relatives) and Crenarchaeota (comprising the relatively tight clustering of extremely thermophilic archaebacteria, whose general phenotype appears to resemble most the ancestral phenotype of the Archaea.