检索结果-内蒙古大学图书馆

Learning to recognize hand-held objects from scratch 17th

学校读者我要写书评

暂无评论

Learning to recognize hand-held objects from scratch

17th Pacific-Rim Conference on Multimedia, PCM 2016

作者： Li, Xue Jiang, Shuqiang Lv, Xiong Chen, Chengpeng College of Information Science and Engineering Shandong University of Science and Technology Qingdao266590 China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China

ISBN: (纸本)9783319488950

Real-life environments are open-ended and dynamic: unlearned information comes over time. These changes of environments ask for the systems to have the ability of self growth. A reasonable solution is to build an intelligent human-computer interaction system to simulate the mind at birth, and then automatically teach it by human. In this work, we present a hand-held object recognition system which could incrementally enhance its recognition ability from beginning during the interaction with humans. Automatically capturing the images of handheld objects and the voice of users, our system could refer the interacting person as a strong teacher. This allows the system to learn from scratch and to learn new concepts one after another like humans. Although our system is implemented on hand-held recognition scenario, we also implement experiments on ImageNet dataset to validate the effectiveness of our system. Experimental results illustrate its performance. © Springer International Publishing AG 2016.

关键词： Object recognition

A simple and efficient fusion framework for surveillance images

学校读者我要写书评

暂无评论

Naukovyi Visnyk Natsionalnoho Hirnychoho Universytetu

Naukovyi Visnyk Natsionalnoho Hirnychoho Universytetu 2016年第6期 149-156页

作者： Chen, Lili Guo, Hongjun Laboratory of Intelligent Information Processing Suzhou University Suzhou China Key Laboratory of Intelligent Computing and Signal Processing of MOE Anhui University Hefei China

Purpose. Aiming at solving the fusion issue of surveillance images, a simple and efficient fusion framework using block compressed sensing sampling (BCSS) is proposed in this paper, which consists of two fusion methods using basic-BCSS and sliding-BCSS respectively. methodology. With the superiority of low sampling ratio and low computational complexity, compressed sensing (CS) theory is widely used in signal processing. The basic-BCSS is a basic version of block based CS, in which the source image is divided into distinct blocks, and the sliding-BCSS is a modified version of basic-BCSS proposed for the first time, in which the image is divided into small sliding blocks for each pixel with appropriate padding. The basic idea of the fusion framework is to select the blocks or pixels with greater L2-norm of the BCSS measurement outputs of the divided blocks in spatial domain. findings. The fusion framework is tested on three pairs of grayscale surveillance images, including infrared and visible images, and millimeter-wave and visible images, and compared with several traditional fusion methods. Experimental results demonstrate that the proposed fusion framework can significantly improve the fusion quality and speed simultaneously. Originality. A simple and efficient fusion framework using BCSS in spatial domain is proposed for the first time. Practical value. It has a certain practical meaning for real-time surveillance applications. © Lili Chen, Hongjun Guo, 2016.

关键词： Image fusion

Graphene based functional devices: A short review

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Wang, Rong Jiang, Lijun Ren, Xingang Yan, Ze Sha, Wei E.I. Shan, Guangcun Department of Electrical and Electronic Engineering University of Hong Kong Pokfulam Road Hong Kong Key Laboratory of Intelligent Computing & Signal Processing Ministry of Education Anhui University Hefei230039 China School of Instrumentation Science & Opto-electronics Engineering Beihang University Beijing100191 China Key Laboratory of Micro-nano Electronic Devices and Smart Systems of Zhejiang Province College of Information Science & Electronic Engineering Zhejiang University Hangzhou310027 China School of Instrumentation Science & Opto-electronics Engineering Beihang University Beijing100191 China California NanoSystem Institute Department of Chemistry & Biochemistry University of California Los AngelesCA90095 United States

Graphene is an ideal 2D material system bridging electronic and photonic devices. It also breaks the fundamental speed and size limits by electronics and photonics, respectively. Graphene offers multiple functions of signal transmission, emission, modulation, and detection in a broad band, high speed, compact size, and low loss. Here, we will have a brief view of graphene based functional devices at microwave, terahertz, and optical frequencies. Their fundamental physics and computational models will be discussed as well. Copyright © 2018, The Authors. All rights reserved.

关键词： Graphene

Automatic cross-lingual similarization of dependency grammars for tree-based machine translation

学校读者我要写书评

暂无评论

Automatic cross-lingual similarization of dependency grammar...

2016 Conference on Empirical Methods in Natural Language processing, EMNLP 2016

作者： Jiang, Wenbin Zhang, Wen Xu, Jinan Cai, Rangjia Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China School of Computer and Information Technology Beijing Jiaotong University China Research Center of Tibetan Information Qinghai Normal University China

ISBN: (纸本)9781945626258

Structural isomorphism between languages benefits the performance of cross-lingual applications. We propose an automatic algorithm for cross-lingual similarization of dependency grammars, which automatically learns grammars with high cross-lingual similarity. The algorithm similarizes the annotation styles of the dependency grammars for two languages in the level of classification decisions, and gradually improves the cross-lingual similarity without losing linguistic knowledge resorting to iterative cross-lingual cooperative learning. The dependency grammars given by cross-lingual similarization have much higher cross-lingual similarity while maintaining non-triviality. As applications, the cross-lingually similarized grammars significantly improve the performance of dependency tree-based machine translation. © 2016 Association for Computational Linguistics

关键词： Machine translation

Flexible and Creative Chinese Poetry Generation Using Neural Memory

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Zhang, Jiyuan Feng, Yang Wang, Dong Wang, Yang Abel, Andrew Zhang, Shiyue Zhang, Andi Tsinghua University Riit China Peking University Shool of Software & Microelectronics China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Cas Huilan Limited Beijing China Beijing University of Posts and Telecommunications China Xi'an Jiaotong-Liverpool University China

It has been shown that Chinese poems can be successfully generated by sequence-to-sequence neural models, particularly with the attention mechanism. A potential problem of this approach, however, is that neural models can only learn abstract rules, while poem generation is a highly creative process that involves not only rules but also innovations for which pure statistical models are not appropriate in principle. This work proposes a memory-augmented neural model for Chinese poem generation, where the neural model and the augmented memory work together to balance the requirements of linguistic accordance and aesthetic innovation, leading to innovative generations that are still rule-compliant. In addition, it is found that the memory mechanism provides interesting flexibility that can be used to generate poems with different styles. Copyright © 2017, The Authors. All rights reserved.

关键词：

Live cell single molecule-guided Bayesian localization super resolution microscopy

学校读者我要写书评

暂无评论

细胞研究（英文版） 2017年第5期27卷 713-716页

作者： Mingshu Zhang Wenting He Zhiyong Liu Fa Zhang Jennifer Lippincott-Schwartz Key Laboratory of RNA Biology Institute of Biophysics Chinese Academy of Sciences Beijing 100101 China Beijing Key Laboratory of Noncoding RNA Institute of Biophysics Chinese Academy of Sciences Beijing 100101 China Key Lab of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing 100190 China Janelia Research Campus Howard Hughes Medical Institute Ashburn Virginia 20147 USA

Dear Editor,Many current super resolution (SR) microscopic techniques [1-6] have been successfully applied to image cellular dynamics in living cells,but their applications have remained technically *** cell stimulated emission depletion (STED)/reversible saturable optical linear fluorescence transitions (RESOLFT) microscopy and structured illumination microscopy (SIM)/nonlinear SIM require sophisticated expensive optical setups and specialized expertise for accurate optical *** cell photo-activated localization microscopy (PALM)/stochastic optical reconstruction microscopy (STORM)use less complicated setup;however,a scientific complementary metal-oxide-semiconductor (sCMOS) camera,whose pixel-dependent noise must be characterized and calibrated before use [7],is required for extremely high acquisition speed over tens of thousands of ***,wide field-based SR microscopies have been developed to improve temporal resolution using much fewer time-lapse images (hundreds to thousands) than PALM/STORM [8,9].One of them,Bayesian analysis of the blinking and bleaching (3B) [8],offers enormous potential to resolve ultrastructure and fast cellular dynamics in living cells beyond the diffraction *** its potential,3B analysis is impractical when imaging the nanoscale dynamics in large fields of view over long time periods,as the calculation is extremely time-consuming [8] and/or the analysis consumes large amounts of web resources [10].Another major problem of 3B imaging is the artificial thinning and thickening of structures both in simulated image data and in experimental data [8,10].

关键词： @@

Hyperspectral light field stereo matching

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Zhu, Kang Xue, Yujia Fu, Qiang Kang, Sing Bing Chen, Xilin Yu, Jingyi School of Information Science and Technology ShanghaiTech University Shanghai China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China Plex-VR Inc. Shanghai Microsoft Research RedmondWA United States Department of Computer and Information Sciences University of Delaware NewarkDE United States

In this paper, we describe how scene depth can be extracted using a hyperspectral light field capture (H-LF) system. Our H-LF system consists of a 5 × 6 array of cameras, with each camera sampling a different narrow band in the visible spectrum. There are two parts to extracting scene depth. The first part is our novel cross-spectral pairwise matching technique, which involves a new spectral-invariant feature descriptor and its companion matching metric we call bidirectional weighted normalized cross correlation (BWNCC). The second part, namely, H-LF stereo matching, uses a combination of spectral-dependent correspondence and defocus cues that rely on BWNCC. These two new cost terms are integrated into a Markov Random Field (MRF) for disparity estimation. Experiments on synthetic and real H-LF data show that our approach can produce high-quality disparity maps. We also show that these results can be used to produce the complete plenoptic cube in addition to synthesizing all-focus and defocused color images under different sensor spectral responses. Copyright © 2017, The Authors. All rights reserved.

关键词： Cameras

Memory-augmented neural machine translation

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Feng, Yang Zhang, Shiyue Zhang, Andi Wang, Dong Abel, Andrew Center for Speech and Language Technologies RIIT Tsinghua University Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing University of Posts and Telecommunications China Huilan Corporation Beijing China Xi'An Jiaotong Liverpool-University Suzhou China

Neural machine translation (NMT) has achieved notable success in recent times, however it is also widely recognized that this approach has limitations with handling infrequent words and word pairs. This paper presents a novel memory-augmented NMT (M-NMT) architecture, which stores knowledge about how words (usually infrequently encountered ones) should be translated in a memory and then utilizes them to assist the neural model. We use this memory mechanism to combine the knowledge learned from a conventional statistical machine translation system and the rules learned by an NMT system, and also propose a solution for out-of-vocabulary (OOV) words based on this framework. Our experiments on two Chinese-English translation tasks demonstrated that the M-NMT architecture outperformed the NMT baseline by 9.0 and 2.7 BLEU points on the two tasks, respectively. Additionally, we found this architecture resulted in a much more effective OOV treatment compared to competitive methods. Copyright © 2017, The Authors. All rights reserved.

关键词： Neural machine translation

学校读者我要写书评

暂无评论

多组学基因大数据与精准医学

多组学基因大数据与精准医学

浙江省生物信息学学会第五届学术年会暨大数据时代精准医学国际学术研讨会

作者： Yi Zhao Bioinformatics Research Group Key Laboratory of Intelligent Information ProcessingAdvanced Computing Research LaboratoryInstitute of Computing TechnologyChineseAcademy of Sciences LuoYang Branch of Institute of Computing Technology Chinese Academy of Sciences

伴随着高通量测序技术等生物技术的发展,基因组、转录组、表观组、蛋白组、代谢组、微生物组、暴露组等多组学测序数据可以在几天之内迅速产出,并被广泛应用于精准医学的研究和临床应用,这些庞大而复杂的数据不但构成了精确医学所需要... 详细信息

伴随着高通量测序技术等生物技术的发展,基因组、转录组、表观组、蛋白组、代谢组、微生物组、暴露组等多组学测序数据可以在几天之内迅速产出,并被广泛应用于精准医学的研究和临床应用,这些庞大而复杂的数据不但构成了精确医学所需要的关键信息,而且也是该领域内增长最快的数据类型。当前,精准医学计划在多个国家陆续启动,各国科学家们正致力于收集十万、百万人群的基因组信息用来构建权威、且全新的人口健康大数据知识体系。在这样的时代背景下,作为人口大国,中国精准医学迫切的需要建立属于中国人自己的基因数据库。2016年,中国科技部将"精准医学研究"列为重大研究计划,中科院也于2016年启动中国人群精准医学研究计划,4年内预计完成4000志愿者的DNA样本和多种表现型数据的采集。对此,我们可以预见,随着大队列研究的开展,以基因组为代表的多组学大数据仍将会不断产生,但是这些组学大数据本身并不能产生价值,其价值在于与疾病表型等其他信息进行关联分析、挖掘并转换为知识,才能进一步被应用于医疗及健康决策。因此,促进大数据到知识的转化,构建基因大数据研究与应用的一体化平台体系,才是精准医学到临床的关键一步,其将会被应用于疾病的早期筛查、分子分型、个体化治疗、疗效预测及监控等多个临床决策场景。多组学精准医学大数据的发展和研究也面临着一些挑战,例如:数据共享与安全隐私问题、多维度组学数据的整合分析问题、生物信息数据分析方法的标准化以及解读规范化等等问题。因此,需要更多国家政策的支持、需要IT及人工智能等领域的学者贡献力量、需要各个领域的专家来制定行业标准,最终推动精准医疗向自动化,智能化,标准化,简单化方向发展。

关键词：精准医学多组学大数据