|
|
|
题名
|
作者
|
年代
|
出处
|
被引量
|
| 1 | Salient object detection: A survey显示文摘Detecting and segmenting salient objects from natural scenes, often referred to as salient object detection, has attracted great interest in computer vision. While many models have been proposed and several applications have emerged, a deep understanding of achievements and issues remains lacking. We aim to provide a comprehensive review of recent progress in salient object detection and situate this field among other closely related areas such as generic scene segmentation, object proposal generation, and saliency for fixation prediction. Covering 228 publications, we survey i) roots, key concepts, and tasks, ii) core techniques and main modeling trends, and iii) datasets and evaluation metrics for salient object detection. We also discuss open problems such as evaluation metrics and dataset bias in model performance, and suggest future research directions. | Ali Borji Ming-Ming Cheng Qibin Hou Huaizu Jiang Jia Li | 2019 | Computational Visual Media2019，5，2： | 40 |
| 2 | Tourette syndrome associated with attention deficit hyperactivity disorder: The impact of tics and psychopharmacological treatment options显示文摘Tourette syndrome(TS) is a neurodevelopmental disorder characterized by multiple chronic motor and vocal tics beginning in childhood. Several studies describe the association between TS and attention deficit hyperactivity disorder(ADHD). Fifty percent of children diagnosed with ADHD have comorbid tic disorder. ADHD related symptoms have been reported in 35% to 90% of children with TS. Since ADHD is the most prevalent comorbid condition with TS and those with concomitant TS and ADHD present with considerable psychosocial and behavioral impairments, it is essential for clinicians to be familiar with these diagnoses and their management. This paper highlights the association between treating ADHD with stimulants and the development of tic disorders. The two cases discussed underscore the fact that children with TS may present with ADHD symptomatology prior to the appearance of any TS related symptoms. Appropriate management of TS in a patient diagnosed with ADHD can lead to quality of life improvements and a reduction in psychosocial impairments. | Olumide O Oluwabusi Susan Parke Paul J Ambrosini | 2016 | World Journal of Clinical Pediatrics2016，5，1： | 27 |
| 3 | Attention mechanisms in computer vision:A survey显示文摘Humans can naturally and effectively find salient regions in complex scenes.Motivated by this observation,attention mechanisms were introduced into computer vision with the aim of imitating this aspect of the human visual system.Such an attention mechanism can be regarded as a dynamic weight adjustment process based on features of the input image.Attention mechanisms have achieved great success in many visual tasks,including image classification,object detection,semantic segmentation,video understanding,image generation,3D vision,multimodal tasks,and self-supervised learning.In this survey,we provide a comprehensive review of various attention mechanisms in computer vision and categorize them according to approach,such as channel attention,spatial attention,temporal attention,and branch attention;a related repository http://gffzz188fe103f8f1460asfoxx5995kxpp66vx.ffgz.tsg.suse.edu.cn/MenghaoG uo/Awesome-Vision-Attentions is dedicated to collecting related work.We also suggest future directions for attention mechanism research. | Meng-Hao Guo Tian-Xing Xu Jiang-Jiang Liu Zheng-Ning Liu Peng-Tao Jiang Tai-Jiang Mu Song-Hai Zhang Ralph R.Martin Ming-Ming Cheng Shi-Min Hu | 2022 | Computational Visual Media2022，8，3： | 27 |
| 4 | 基于BERT与Bi-LSTM融合注意力机制的中医病历文本的提取与自动分类显示文摘中医逐渐成为热点,中医病历文本中包含着巨大而宝贵的医疗信息。而在中医病历文本挖掘和利用方面,一直面临中医病历文本利用率低、抽取有效信息并对信息文本进行分类的难度大的问题。针对这一问题,研究一种对中医病历文本的提取与自动分类的方法具有很大的临床价值。文中尝试提出一种基于BERT+Bi-LSTM+Attention融合的病历短文本分类模型。使用BERT预处理获取短文本向量作为模型输入,对比BERT与word2vec模型的预训练效果,对比Bi-LSTM+Attention和LSTM模型的效果。实验结果表明,BERT+Bi-LSTM+Attention融合模型在中医病历文本的提取和分类方面达到了最高的AverageF1值(即89.52%)。通过对比发现,BERT较word2vec模型的预训练效果有显著的提升,且Bi-LSTM+Attention模型较LSTM模型的效果有显著的提升,因此提出的BERT+Bi-LSTM+Attention融合模型在病历文本抽取与分类上有一定的医学价值。 | 杜琳 曹东 林树元 瞿溢谦 叶辉 | 2020 | 计算机科学2020，47，S02： | 23 |
| 5 | Socio-cognitive Approach to Pragmatics显示文摘Communication is not as smooth a process as current pragmatic theories depict it.In Rapaport's words 'We almost always fail .Yet we almost always nearly succeed: This is the paradox of communication(Rapaport,2003:402).' This paper claims that there is a need for an approach that is able to explain this 'bumpy road' by analyzing both the positive and negative features of the communicative process.The paper presents a socio-cognitive approach(SCA) to pragmatics that takes into account both the societal and individual factors including cooperation and egocentrism that,as claimed here,are not antagonistic phenomena in interaction.This approach is considered an alternative to current theories of pragmatics that do not give an adequate account of what really happens in the communicative process.They consider communication an idealistic,cooperation-based,context-dependent process in which speakers are supposed to carefully construct their utterances for the hearer taking into account all contextual factors and hearers do their best to figure out the intentions of the speakers.This approach relies mainly on the positive features of communication including cooperation,rapport and politeness while almost completely ignores the untidy,trial-and-error nature of communication and the importance of prior contexts captured in the individual use of linguistic units.The overemphasis on cooperative,societal,contextual factors has led to disregard individual factors such as egocentrism and salience that are as important contributors to the communicative process as cooperation,context and rapport.The socio-cognitive approach is presented as a theoretical framework to incorporate and reconcile two seemingly antagonistic sides of the communicative process and explain the dynamic interplay of prior and actual situational contexts. | Istvan Kecskes | 2010 | 外国语2010，33，5： | 16 |
| 6 | Access,Activation,and Overlap:Focusing on the Differential显示文摘As a long term goal,Cognitive Grammar envisages a unified account of linguistic structure,processing,and discourse.A key requirement is that it properly reflect the dynamicity of language,in all its aspects(including grammar).Consisting as it does in patterns of activity,language is something that happens: it unfolds through time,and how it does so is essential for its characterization.Language structure involves the complex interplay of serial and hierarchical organization.Linguistic theory has generally overemphasized the latter,as witnessed by the predominant role of grammatical constituency based on the metaphor of composition('building' a complex structure out of smaller 'pieces').More promising for a unified treatment of structure,processing,and discourse is an alternative metaphor,in which a 'moving window' of attention allows serial access to complex conceptions that are mostly already in place.Serial processing proceeds concurrently on different time scales,in windows of different duration.This provides a dynamic basis for hierarchical organization.Constituency arises when conceptions accessed in windows on one time scale figure in a distinct and more elaborate conception(with its own,emergent properties) accessed in a single window on a larger time scale.Based on the alternative metaphor,an approach is outlined that features serial access,windows of attention,levels of activation,and conceptual overlap.It is then applied,in exploratory fashion,to a number of grammatical phenomena.The 'access-and-activation' approach directly accommodates reference point relationships,the basis for possessive and topic constructions.It lends itself quite well to the characterization of informational focus,marked in English by unreduced stress.The requisite descriptive notions are then extended and adapted for ellipsis,which has rather different properties.The same notions prove useful for handling problematic aspects of coordinate constructions.Finally,they provide a straightforward means of dealing with 'zero-anaphora'.The access-and-activation approach does not entail any modification of Cognitive Grammar.It simply makes explicit certain descriptive options afforded by its central claims. | Ronald W.Langacker | 2012 | 外国语2012，35，1： | 15 |
| 7 | Hybrid first and second order attention Unet for building segmentation in remote sensing images显示文摘Recently,building segmentation(BS)has drawn significant attention in remote sensing applications.Convolutional neural networks(CNNs)have become the mainstream analysis approach in this field owing to their powerful representative ability.However,owing to the variation in building appearance,designing an effective CNN architecture for BS still remains a challenging task.Most of CNN-based BS methods mainly focus on deep or wide network architectures,neglecting the correlation among intermediate features.To address this problem,in this paper we propose a hybrid first and second order attention network(HFSA)that explores both the global mean and the inner-product among different channels to adaptively rescale intermediate features.As a result,the HFSA can not only make full use of first order feature statistics,but also incorporate the second order feature statistics,which leads to more representative feature.We conduct a series of comprehensive experiments on three widely used aerial building segmentation data sets and one satellite building segmentation data set.The experimental results show that our newly developed model achieves better segmentation performance over state-of-the-art models in terms of both quantitative and qualitative results. | Nanjun HE Leyuan FANG Antonio PLAZA | 2020 | Science China(Information Sciences)2020，63，4： | 15 |
| 8 | 基于z指数的AAS高关注度学科研究主题识别显示文摘为了准确、有效地识别学科研究热点与前沿主题,文章选择Altmetric.com平台17个指标加权综合的Alt-metrics Attention Score(以下简称AAS)值,基于z指数方法识别高关注度研究主题.选择SSCI数据库中情报学5种优秀期刊2018年的论文为样本,获取这些论文的AAS值,将AAS值替代z指数中的被引频次,构建zt指数模型识别高AAS关注度研究主题,构建平均数基准线分类模型对识别结果进行细分.识别出情报学的16个高关注度研究主题,并细分潜力类、突现类、核心类、边缘类4类学科研究内容,发现本文构建的zt指数模型和高关注度学科研究主题分类模型是可行、有效的. | 牌艳欣 李长玲 刘运梅 | 2019 | 情报资料工作2019，40，6： | 11 |
| 9 | BING: Binarized normed gradients for objectness estimation at 300fps显示文摘Training a generic objectness measure to produce object proposals has recently become of significant interest. We observe that generic objects with well-defined closed boundaries can be detected by looking at the norm of gradients, with a suitable resizing of their corresponding image windows to a small fixed size. Based on this observation and computational reasons, we propose to resize the window to 8 × 8 and use the norm of the gradients as a simple 64 D feature to describe it, for explicitly training a generic objectness measure. We further show how the binarized version of this feature, namely binarized normed gradients(BING), can be used for efficient objectness estimation, which requires only a few atomic operations(e.g., add, bitwise shift, etc.). To improve localization quality of the proposals while maintaining efficiency, we propose a novel fast segmentation method and demonstrate its effectiveness for improving BING's localization performance, when used in multithresholding straddling expansion(MTSE) postprocessing. On the challenging PASCAL VOC2007 dataset, using 1000 proposals per image and intersectionover-union threshold of 0.5, our proposal method achieves a 95.6% object detection rate and 78.6% mean average best overlap in less than 0.005 second per image. | Ming-Ming Cheng Yun Liu Wen-Yan Lin Ziming Zhang Paul L.Rosin Philip H.S.Torr | 2019 | Computational Visual Media2019，5，1： | 11 |
| 10 | 基于Multi-head Attention和Bi-LSTM的实体关系分类显示文摘关系分类是自然语言处理领域的一项重要任务,能够为知识图谱的构建、问答系统和信息检索等提供技术支持.与传统关系分类方法相比较,基于神经网络和注意力机制的关系分类模型在各种关系分类任务中都获得了更出色的表现.以往的模型大多采用单层注意力机制,特征表达相对单一.因此本文在已有研究基础上,引入多头注意力机制(Multi-head attention),旨在让模型从不同表示空间上获取关于句子更多层面的信息,提高模型的特征表达能力.同时在现有的词向量和位置向量作为网络输入的基础上,进一步引入依存句法特征和相对核心谓词依赖特征,其中依存句法特征包括当前词的依存关系值和所依赖的父节点位置,从而使模型进一步获取更多的文本句法信息.在SemEval-2010 任务8 数据集上的实验结果证明,该方法相较之前的深度学习模型,性能有进一步提高. | 刘峰 高赛 于碧辉 郭放达 | 2019 | 计算机系统应用2019，28，6： | 11 |
| 11 | Review of the evidence for the management of co-morbid Tic disorders in children and adolescents with attention deficit hyperactivity disorder显示文摘Attention deficit hyperactivity disorder(ADHD) is the most common neurodevelopmental disorder in children and adolescents, with prevalence ranging between 5% and 12% in the developed countries. Tic disorders(TD) are common co-morbidities in paediatric ADHD patients with or without pharmacotherapy treatment. There has been conflicting evidence of the role of psychostimulants in either precipitating or exacerbating TDs in ADHD patients. We carried out a literature review relating to the management of TDs in children and adolescents with ADHD through a comprehensive search of MEDLINE, EMBASE, CINAHL and Cochrane databases. No quantitative synthesis(meta-analysis) was deemed appropriate. Metaanalysis of controlled trials does not support an association between new onset or worsening of tics and normal doses of psychostimulant use. Supratherapeutic doses of dextroamphetamine have been shown to exacerbate TD. Most tics are mild or moderate and respond to psychoeducation and behavioural management. Level A evidence support the use of alpha adrenergic agonists, including Clonidine and Guanfacine, reuptake noradrenenaline inhibitors(Atomoxetine) and stimulants(Methylphenidate and Dexamphetamines) for the treatment of Tics and comorbid ADHD. Priority should be given to the management of co-morbid Tourette's syndrome(TS) or severely disabling tics in children and adolescents with ADHD. Severe TDs may require antipsychotic treatment. Antipsychotics, especially Aripiprazole, are safe and effective treatment for TS or severe Tics, but they only moderately control the co-occurring ADHD symptomatology. Short vignettes of different common clinical scenarios are presented to help clinicians determine the most appropriate treatment to consider in each patient presenting with ADHD and co-morbid TDs. | Michael O Ogundele Hani F Ayyash | 2018 | World Journal of Clinical Pediatrics2018，7，1： | 10 |
| 12 | 宜宾油樟对小麦的化感作用研究显示文摘采用生物测定法研究了宜宾油樟根、茎、叶水浸提液对小麦种子萌发及幼苗生长的化感作用.结果表明,油樟对小麦的化感作用和供体的不同器官、浸提液浓度以及受体的不同发育阶段有密切关系.油樟浸提液主要延迟小麦种子的萌发.随浸提液浓度的增大,小麦种子的最终发芽率、发芽速率、幼苗叶绿素含量呈'降-升-降'的趋势变化,而幼苗高度、根长呈'升-降'的趋势变化.化感强度综合效应是叶>根>茎,提示人们在小麦生产区要注意油樟种植的密度. | 邓骛远 罗通 彭铄钧 | 2009 | 四川大学学报(自然科学版)2009，46，6： | 10 |
| 13 | Relation Classification via Recurrent Neural Network with Attention and Tensor Layers显示文摘Relation classification is a crucial component in many Natural Language Processing(NLP) systems. In this paper, we propose a novel bidirectional recurrent neural network architecture(using Long Short-Term Memory,LSTM, cells) for relation classification, with an attention layer for organizing the context information on the word level and a tensor layer for detecting complex connections between two entities. The above two feature extraction operations are based on the LSTM networks and use their outputs. Our model allows end-to-end learning from the raw sentences in the dataset, without trimming or reconstructing them. Experiments on the SemEval-2010 Task 8dataset show that our model outperforms most state-of-the-art methods. | Runyan Zhang Fanrong Meng Yong Zhou Bing Liu | 2018 | Big Data Mining and Analytics2018，1，3： | 9 |
| 14 | 基于ERNIE2.0-BiLSTM-Attention的隐式情感分析方法显示文摘隐式情感分析作为自然语言处理领域的子任务,因不具备显式情感词作为情感线索,使得传统文本情感分析方法不再有效.本文旨在使用深度学习方法进行文本的隐式情感分析,根据文本隐式情感极性与句中实体、上下文语境、外部知识相关的特点,本文提出一种基于ERNIE2.0-BiLSTM-Attention(EBA)的隐式情感分析方法,能够较好捕捉隐式情感句的语义及上下文信息,有效提升隐式情感的识别能力,最后在SMP2019公开数据集上取得较好分类效果,分类模型准确率达到82.3%. | 黄山成 韩东红 乔百友 吴刚 王国仁 | 2021 | 小型微型计算机系统2021，42，12： | 9 |
| 15 | Brain dynamic mechanisms on the visual attention scale with Chinese characters cues显示文摘The temporal dynamics in brain evoked by the scale of visual attention with the cues of Chinese characters were studied by recording event-related potentials (ERPs). With the fixed orientation of visual attention, 14 healthy young participants performed a search task in which the search array was preceded by Chinese characters cues, '大, 中, 小' (large, medium, small). 128 channels scalp ERPs were recorded to study the role of visual attention scale played in the visual spatial attention. The results showed that there was no significant difference in the ERP components evoked by the three Chinese characters cues except the in-feroposterior N2 latency. The targets evoked P2, N2 amplitudes and latency have significant differences with the different cues of large, middle and small, while P1 and N1 components had no significant difference. The results suggested that the processing of scale of visual attention was mainly concerned with P2, N2 components, while the P1, N1 components were mainly related with | GAO Wenbin WEI Jinghan PENG Xiaohu WEI Xing LUO Yuejia | 2002 | Chinese Science Bulletin2002，47，19： | 8 |
| 16 | 水对硅酸盐岩体系部分熔融行为的影响:第二临界端点的重要意义显示文摘水对硅酸盐岩体系的许多物理—化学行为有着非常重大的影响。具体对部分熔融过程来说,水可以显著地降低熔融温度、改变熔体性质、影响微量元素在固—液相之间的分配。近年来,科学家们就大量水对硅酸盐岩体系的部分熔融过程的影响进行了许多的高压实验,他们主要关注第二临界端点对熔融过程的重要作用:第二临界端点的出现极大地改变着部分熔融过程中的基本相关系。本文主要针对这些高压实验研究做一总结,并对未来研究方向做一初步探讨。 | 刘曦 张立飞 HACK C Alistair 郑海飞 胡晓敏 常琳琳 何强 | 2009 | 岩石学报2009，25，12： | 8 |
| 17 | Research progress on Drosophila visual cognition in China显示文摘Visual cognition,as one of the fundamental aspects of cognitive neuroscience,is generally associated with high-order brain functions in animals and human.Drosophila,as a model organism,shares certain features of visual cognition in common with mammals at the genetic,molecular,cellular,and even higher behavioral levels.From learning and memory to decision making,Drosophila covers a broad spectrum of higher cognitive behaviors beyond what we had expected.Armed with powerful tools of genetic manipulation in Drosophila,an increasing number of studies have been conducted in order to elucidate the neural circuit mechanisms underlying these cognitive behaviors from a genes-brain-behavior perspective.The goal of this review is to integrate the most important studies on visual cognition in Drosophila carried out in China's Mainland during the last decade into a body of knowledge encompassing both the basic neural operations and circuitry of higher brain function in Drosophila.Here,we consider a series of the higher cognitive behaviors beyond learning and memory,such as visual pattern recognition,feature and context generalization,different feature memory traces,salience-based decision,attention-like behavior,and cross-modal leaning and memory.We discuss the possible general gain-gating mechanism implementing by dopamine-mushroom body circuit in fly's visual cognition.We hope that our brief review on this aspect will inspire further study on visual cognition in flies,or even beyond. | GUO AiKe1,2,ZHANG Ke1,PENG YueQin1 & XI Wang1 1 Institute of Neuroscience,State Key Laboratory of Neuroscience,Shanghai Institutes for Biological Sciences,Chinese Academy of Sciences,Shanghai 200031,China 2 State Key Laboratory of Brain and Cognitive Science,Institute of Biophysics,Chinese Academy of Sciences,Beijing 100101,China | 2010 | Science China(Life Sciences)2010，53，3： | 6 |
| 18 | 互联网新闻敏感信息识别方法的研究显示文摘敏感信息识别是净化互联网环境的关键,在当今信息爆炸的时代,人们每天都要从互联网中获得大量信息,如何过滤大量信息中的敏感信息对整个社会安定和谐有着重要的意义.现有的方法主要是基于敏感关键词的方法进行过滤,需要不断更新迭代敏感关键词,泛化性弱,本文中使用基于预训练模型的深度学习方法可以学习到互联网新闻文本中更深层的语义信息,进而更有效的识别和过滤敏感信息,泛化性强,但是只使用深度学习方法会一定程度上的损失敏感关键词特征.本文首次将传统的敏感关键词方法与深度学习方法相结合应用于互联网敏感信息识别,提出了一种融合敏感关键词特征的模型Mer-HiBert.实验结果表明,与之前的敏感关键词方法以及深度学习模型相比,模型的性能有进一步提高. | 李姝 张祥祥 于碧辉 于金刚 | 2021 | 小型微型计算机系统2021，42，4： | 6 |
| 19 | Irregular scene text detection via attention guided border labeling显示文摘Scene text detection plays an important role in many computer vision applications. With the help of recent deep learning techniques, multi-oriented text detection that was considered to be quite challenging has been solved to some extent. However, most existing methods still perform poorly for curved text detection, mainly due to the limitation of their text representations(e.g., horizontal boxes, rotated rectangles or quadrangles). To solve this problem, we propose a novel method to detect irregular scene texts based on instance-aware segmentation. The key idea is to design an attention guided semantic segmentation model to precisely label the weighted borders of text regions. Experiments conducted on several widely-used benchmarks demonstrate that our method achieves superior results on curved text datasets(i.e., with F-score80.1% and 78.8% for the CTW1500 and Total-Text, respectively) and obtains comparable performance on multi-oriented text datasets compared to the state-of-the-art approaches. | Jie CHEN Zhouhui LIAN Yizhi WANG Yingmin TANG Jianguo XIAO | 2019 | Science China(Information Sciences)2019，62，12： | 6 |
| 20 | Does mindfulness meditation improve attention in attention deficit hyperactivity disorder?显示文摘Attention deficit hyperactivity disorder(ADHD) manifests by high levels of inattention, impulsiveness and hyperactivity. ADHD starts in childhood and results in impairments that continue into adulthood. While hyperactivity declines over time, inattention and executive function difficulties persist, leading to functional deficits. Adolescents and adults with ADHD have pervasive impairment in interpersonal and family relationships. They may develop addiction, delinquent behavior and comorbid psychiatric disorders. Despite advances in diagnosis and treatment, persistent residual symptoms are common, highlighting the need for novel treatment strategies. Mindfulness training, derived from Eastern meditation practices, may improve self-regulation of attention. It may also be a useful strategy to augment standard ADHD treatments and may be used as a potential tool to reduce impairments in patients with residual symptoms of ADHD. Clinically, this would manifest by an increased ability to suppress task-unrelated thoughts and distractions resulting in improved attention, completion of tasks and potential improvement in occupational and social function. | Vania Modesto-Lowe Pantea Farahmand Margaret Chaplin Lauren Sarro | 2015 | World Journal of Psychiatry2015，5，4： | 5 |
您所在的机构暂未开通此部分的功能，请联系管理员